dan-latn-100mb-100mb_seed3407/checkpoint-33000/trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.75643167390578,
  "eval_steps": 3000,
  "global_step": 33000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 10.69201889038086,
      "epoch": 0.000417641162712997,
      "grad_norm": 14.0,
      "learning_rate": 2e-06,
      "loss": 10.8059,
      "mean_token_accuracy": 0.0,
      "num_tokens": 9371.0,
      "step": 5
    },
    {
      "entropy": 10.691970920562744,
      "epoch": 0.000835282325425994,
      "grad_norm": 12.875,
      "learning_rate": 4.5e-06,
      "loss": 10.7231,
      "mean_token_accuracy": 0.0003899427130818367,
      "num_tokens": 18155.0,
      "step": 10
    },
    {
      "entropy": 10.691130447387696,
      "epoch": 0.001252923488138991,
      "grad_norm": 9.125,
      "learning_rate": 7e-06,
      "loss": 10.4903,
      "mean_token_accuracy": 0.024901279783807694,
      "num_tokens": 27776.0,
      "step": 15
    },
    {
      "entropy": 10.681210708618163,
      "epoch": 0.001670564650851988,
      "grad_norm": 6.34375,
      "learning_rate": 9.5e-06,
      "loss": 10.1859,
      "mean_token_accuracy": 0.04129145070910454,
      "num_tokens": 37166.0,
      "step": 20
    },
    {
      "entropy": 10.632552528381348,
      "epoch": 0.002088205813564985,
      "grad_norm": 4.21875,
      "learning_rate": 1.2e-05,
      "loss": 9.8516,
      "mean_token_accuracy": 0.04321358092129231,
      "num_tokens": 46667.0,
      "step": 25
    },
    {
      "entropy": 10.590499591827392,
      "epoch": 0.002505846976277982,
      "grad_norm": 3.296875,
      "learning_rate": 1.4500000000000002e-05,
      "loss": 9.6786,
      "mean_token_accuracy": 0.04258152581751347,
      "num_tokens": 55738.0,
      "step": 30
    },
    {
      "entropy": 10.598013877868652,
      "epoch": 0.002923488138990979,
      "grad_norm": 2.78125,
      "learning_rate": 1.7000000000000003e-05,
      "loss": 9.6768,
      "mean_token_accuracy": 0.04020156487822533,
      "num_tokens": 65303.0,
      "step": 35
    },
    {
      "entropy": 10.575642108917236,
      "epoch": 0.003341129301703976,
      "grad_norm": 2.78125,
      "learning_rate": 1.95e-05,
      "loss": 9.5538,
      "mean_token_accuracy": 0.044753269851207735,
      "num_tokens": 74876.0,
      "step": 40
    },
    {
      "entropy": 10.562861919403076,
      "epoch": 0.003758770464416973,
      "grad_norm": 2.421875,
      "learning_rate": 2.2e-05,
      "loss": 9.515,
      "mean_token_accuracy": 0.04800709597766399,
      "num_tokens": 84683.0,
      "step": 45
    },
    {
      "entropy": 10.554056072235108,
      "epoch": 0.00417641162712997,
      "grad_norm": 2.578125,
      "learning_rate": 2.4500000000000003e-05,
      "loss": 9.4698,
      "mean_token_accuracy": 0.045513217151165006,
      "num_tokens": 94511.0,
      "step": 50
    },
    {
      "entropy": 10.544468021392822,
      "epoch": 0.004594052789842967,
      "grad_norm": 2.46875,
      "learning_rate": 2.7e-05,
      "loss": 9.3749,
      "mean_token_accuracy": 0.056402480229735374,
      "num_tokens": 103820.0,
      "step": 55
    },
    {
      "entropy": 10.37349443435669,
      "epoch": 0.005011693952555964,
      "grad_norm": 2.578125,
      "learning_rate": 2.95e-05,
      "loss": 9.2941,
      "mean_token_accuracy": 0.059941830858588216,
      "num_tokens": 112398.0,
      "step": 60
    },
    {
      "entropy": 10.357733345031738,
      "epoch": 0.005429335115268961,
      "grad_norm": 2.46875,
      "learning_rate": 3.2e-05,
      "loss": 9.2329,
      "mean_token_accuracy": 0.058718277141451836,
      "num_tokens": 121167.0,
      "step": 65
    },
    {
      "entropy": 10.474106597900391,
      "epoch": 0.005846976277981958,
      "grad_norm": 2.34375,
      "learning_rate": 3.4500000000000005e-05,
      "loss": 9.1346,
      "mean_token_accuracy": 0.05946669317781925,
      "num_tokens": 129399.0,
      "step": 70
    },
    {
      "entropy": 10.413960456848145,
      "epoch": 0.006264617440694955,
      "grad_norm": 2.5,
      "learning_rate": 3.7e-05,
      "loss": 9.0661,
      "mean_token_accuracy": 0.05361619554460049,
      "num_tokens": 138512.0,
      "step": 75
    },
    {
      "entropy": 10.455136489868163,
      "epoch": 0.006682258603407952,
      "grad_norm": 2.390625,
      "learning_rate": 3.95e-05,
      "loss": 9.027,
      "mean_token_accuracy": 0.058849292993545535,
      "num_tokens": 147712.0,
      "step": 80
    },
    {
      "entropy": 10.360787677764893,
      "epoch": 0.007099899766120949,
      "grad_norm": 2.375,
      "learning_rate": 4.2000000000000004e-05,
      "loss": 8.8732,
      "mean_token_accuracy": 0.06726518124341965,
      "num_tokens": 156759.0,
      "step": 85
    },
    {
      "entropy": 10.307044124603271,
      "epoch": 0.007517540928833946,
      "grad_norm": 2.328125,
      "learning_rate": 4.45e-05,
      "loss": 8.7741,
      "mean_token_accuracy": 0.06508033163845539,
      "num_tokens": 166395.0,
      "step": 90
    },
    {
      "entropy": 10.328545475006104,
      "epoch": 0.007935182091546943,
      "grad_norm": 2.15625,
      "learning_rate": 4.7000000000000004e-05,
      "loss": 8.6533,
      "mean_token_accuracy": 0.06614198088645935,
      "num_tokens": 174739.0,
      "step": 95
    },
    {
      "entropy": 10.214155673980713,
      "epoch": 0.00835282325425994,
      "grad_norm": 3.484375,
      "learning_rate": 4.9500000000000004e-05,
      "loss": 8.4788,
      "mean_token_accuracy": 0.07055597789585591,
      "num_tokens": 183010.0,
      "step": 100
    },
    {
      "entropy": 10.180351448059081,
      "epoch": 0.008770464416972937,
      "grad_norm": 2.125,
      "learning_rate": 5.2e-05,
      "loss": 8.4699,
      "mean_token_accuracy": 0.06824311949312686,
      "num_tokens": 192589.0,
      "step": 105
    },
    {
      "entropy": 10.146885681152344,
      "epoch": 0.009188105579685933,
      "grad_norm": 2.28125,
      "learning_rate": 5.45e-05,
      "loss": 8.3311,
      "mean_token_accuracy": 0.07665134817361832,
      "num_tokens": 201638.0,
      "step": 110
    },
    {
      "entropy": 10.078996276855468,
      "epoch": 0.00960574674239893,
      "grad_norm": 2.390625,
      "learning_rate": 5.7e-05,
      "loss": 8.2014,
      "mean_token_accuracy": 0.07701764479279519,
      "num_tokens": 209862.0,
      "step": 115
    },
    {
      "entropy": 10.014488697052002,
      "epoch": 0.010023387905111928,
      "grad_norm": 2.015625,
      "learning_rate": 5.9499999999999996e-05,
      "loss": 8.2213,
      "mean_token_accuracy": 0.0736879549920559,
      "num_tokens": 218897.0,
      "step": 120
    },
    {
      "entropy": 9.855342769622803,
      "epoch": 0.010441029067824925,
      "grad_norm": 2.453125,
      "learning_rate": 6.2e-05,
      "loss": 8.1073,
      "mean_token_accuracy": 0.07421992868185043,
      "num_tokens": 227682.0,
      "step": 125
    },
    {
      "entropy": 9.788538360595703,
      "epoch": 0.010858670230537921,
      "grad_norm": 2.21875,
      "learning_rate": 6.450000000000001e-05,
      "loss": 7.9192,
      "mean_token_accuracy": 0.07975538857281209,
      "num_tokens": 236170.0,
      "step": 130
    },
    {
      "entropy": 9.63622751235962,
      "epoch": 0.011276311393250919,
      "grad_norm": 2.25,
      "learning_rate": 6.7e-05,
      "loss": 7.8973,
      "mean_token_accuracy": 0.07390527278184891,
      "num_tokens": 245710.0,
      "step": 135
    },
    {
      "entropy": 9.484860038757324,
      "epoch": 0.011693952555963916,
      "grad_norm": 1.8046875,
      "learning_rate": 6.950000000000001e-05,
      "loss": 7.7354,
      "mean_token_accuracy": 0.07942189872264863,
      "num_tokens": 254751.0,
      "step": 140
    },
    {
      "entropy": 9.272345066070557,
      "epoch": 0.012111593718676913,
      "grad_norm": 2.109375,
      "learning_rate": 7.2e-05,
      "loss": 7.6203,
      "mean_token_accuracy": 0.08180429227650166,
      "num_tokens": 264210.0,
      "step": 145
    },
    {
      "entropy": 9.060223388671876,
      "epoch": 0.01252923488138991,
      "grad_norm": 1.7578125,
      "learning_rate": 7.45e-05,
      "loss": 7.5703,
      "mean_token_accuracy": 0.0850793406367302,
      "num_tokens": 272944.0,
      "step": 150
    },
    {
      "entropy": 8.841773128509521,
      "epoch": 0.012946876044102907,
      "grad_norm": 1.875,
      "learning_rate": 7.7e-05,
      "loss": 7.4347,
      "mean_token_accuracy": 0.08460724949836732,
      "num_tokens": 281311.0,
      "step": 155
    },
    {
      "entropy": 8.642339897155761,
      "epoch": 0.013364517206815904,
      "grad_norm": 1.6484375,
      "learning_rate": 7.950000000000001e-05,
      "loss": 7.3341,
      "mean_token_accuracy": 0.08855588883161544,
      "num_tokens": 289701.0,
      "step": 160
    },
    {
      "entropy": 8.47796277999878,
      "epoch": 0.013782158369528901,
      "grad_norm": 1.1953125,
      "learning_rate": 8.2e-05,
      "loss": 7.3955,
      "mean_token_accuracy": 0.08427253440022468,
      "num_tokens": 299459.0,
      "step": 165
    },
    {
      "entropy": 8.328273963928222,
      "epoch": 0.014199799532241897,
      "grad_norm": 1.5859375,
      "learning_rate": 8.450000000000001e-05,
      "loss": 7.3601,
      "mean_token_accuracy": 0.08567551225423813,
      "num_tokens": 309343.0,
      "step": 170
    },
    {
      "entropy": 8.229494094848633,
      "epoch": 0.014617440694954895,
      "grad_norm": 1.5546875,
      "learning_rate": 8.7e-05,
      "loss": 7.3237,
      "mean_token_accuracy": 0.07697097286581993,
      "num_tokens": 318840.0,
      "step": 175
    },
    {
      "entropy": 8.055557680130004,
      "epoch": 0.015035081857667892,
      "grad_norm": 1.3515625,
      "learning_rate": 8.95e-05,
      "loss": 7.1265,
      "mean_token_accuracy": 0.08917100429534912,
      "num_tokens": 328689.0,
      "step": 180
    },
    {
      "entropy": 8.041841316223145,
      "epoch": 0.01545272302038089,
      "grad_norm": 1.6640625,
      "learning_rate": 9.2e-05,
      "loss": 7.2541,
      "mean_token_accuracy": 0.08379325792193412,
      "num_tokens": 338209.0,
      "step": 185
    },
    {
      "entropy": 7.833249950408936,
      "epoch": 0.015870364183093885,
      "grad_norm": 1.1484375,
      "learning_rate": 9.45e-05,
      "loss": 7.1264,
      "mean_token_accuracy": 0.08980818912386894,
      "num_tokens": 347270.0,
      "step": 190
    },
    {
      "entropy": 7.8264617919921875,
      "epoch": 0.016288005345806884,
      "grad_norm": 1.515625,
      "learning_rate": 9.7e-05,
      "loss": 7.1514,
      "mean_token_accuracy": 0.09062685966491699,
      "num_tokens": 356366.0,
      "step": 195
    },
    {
      "entropy": 7.812281751632691,
      "epoch": 0.01670564650851988,
      "grad_norm": 1.53125,
      "learning_rate": 9.95e-05,
      "loss": 7.1091,
      "mean_token_accuracy": 0.08819653950631619,
      "num_tokens": 365980.0,
      "step": 200
    },
    {
      "entropy": 7.708438777923584,
      "epoch": 0.017123287671232876,
      "grad_norm": 1.4453125,
      "learning_rate": 0.000102,
      "loss": 7.0677,
      "mean_token_accuracy": 0.09280576929450035,
      "num_tokens": 374591.0,
      "step": 205
    },
    {
      "entropy": 7.678690814971924,
      "epoch": 0.017540928833945875,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00010449999999999999,
      "loss": 7.0653,
      "mean_token_accuracy": 0.09315739646553993,
      "num_tokens": 383250.0,
      "step": 210
    },
    {
      "entropy": 7.6943333625793455,
      "epoch": 0.01795856999665887,
      "grad_norm": 1.40625,
      "learning_rate": 0.000107,
      "loss": 6.9805,
      "mean_token_accuracy": 0.08931331932544709,
      "num_tokens": 391502.0,
      "step": 215
    },
    {
      "entropy": 7.598643970489502,
      "epoch": 0.018376211159371866,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001095,
      "loss": 6.901,
      "mean_token_accuracy": 0.09360511749982833,
      "num_tokens": 400225.0,
      "step": 220
    },
    {
      "entropy": 7.616733074188232,
      "epoch": 0.018793852322084866,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000112,
      "loss": 7.1181,
      "mean_token_accuracy": 0.0864988811314106,
      "num_tokens": 410008.0,
      "step": 225
    },
    {
      "entropy": 7.581192922592163,
      "epoch": 0.01921149348479786,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0001145,
      "loss": 7.0294,
      "mean_token_accuracy": 0.09357244744896889,
      "num_tokens": 418979.0,
      "step": 230
    },
    {
      "entropy": 7.522384977340698,
      "epoch": 0.01962913464751086,
      "grad_norm": 1.25,
      "learning_rate": 0.00011700000000000001,
      "loss": 7.0167,
      "mean_token_accuracy": 0.0893492966890335,
      "num_tokens": 428139.0,
      "step": 235
    },
    {
      "entropy": 7.554843139648438,
      "epoch": 0.020046775810223856,
      "grad_norm": 1.59375,
      "learning_rate": 0.00011949999999999999,
      "loss": 6.9224,
      "mean_token_accuracy": 0.09392503276467323,
      "num_tokens": 437113.0,
      "step": 240
    },
    {
      "entropy": 7.548024368286133,
      "epoch": 0.020464416972936852,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000122,
      "loss": 7.1405,
      "mean_token_accuracy": 0.09140172526240349,
      "num_tokens": 446879.0,
      "step": 245
    },
    {
      "entropy": 7.432359743118286,
      "epoch": 0.02088205813564985,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001245,
      "loss": 7.0354,
      "mean_token_accuracy": 0.08848990350961686,
      "num_tokens": 456362.0,
      "step": 250
    },
    {
      "entropy": 7.506717348098755,
      "epoch": 0.021299699298362847,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000127,
      "loss": 6.8268,
      "mean_token_accuracy": 0.09714673981070518,
      "num_tokens": 464930.0,
      "step": 255
    },
    {
      "entropy": 7.499800443649292,
      "epoch": 0.021717340461075842,
      "grad_norm": 1.203125,
      "learning_rate": 0.0001295,
      "loss": 7.0129,
      "mean_token_accuracy": 0.08504997566342354,
      "num_tokens": 474303.0,
      "step": 260
    },
    {
      "entropy": 7.423412275314331,
      "epoch": 0.02213498162378884,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000132,
      "loss": 7.0134,
      "mean_token_accuracy": 0.08980379328131675,
      "num_tokens": 483750.0,
      "step": 265
    },
    {
      "entropy": 7.405795764923096,
      "epoch": 0.022552622786501837,
      "grad_norm": 1.25,
      "learning_rate": 0.00013450000000000002,
      "loss": 6.9026,
      "mean_token_accuracy": 0.09115351662039757,
      "num_tokens": 493325.0,
      "step": 270
    },
    {
      "entropy": 7.417911052703857,
      "epoch": 0.022970263949214833,
      "grad_norm": 1.359375,
      "learning_rate": 0.00013700000000000002,
      "loss": 6.8963,
      "mean_token_accuracy": 0.08686210587620735,
      "num_tokens": 501907.0,
      "step": 275
    },
    {
      "entropy": 7.393596076965332,
      "epoch": 0.023387905111927832,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0001395,
      "loss": 6.9521,
      "mean_token_accuracy": 0.09035036042332649,
      "num_tokens": 511522.0,
      "step": 280
    },
    {
      "entropy": 7.3399505615234375,
      "epoch": 0.023805546274640828,
      "grad_norm": 1.3125,
      "learning_rate": 0.00014199999999999998,
      "loss": 6.8929,
      "mean_token_accuracy": 0.09046279191970825,
      "num_tokens": 521287.0,
      "step": 285
    },
    {
      "entropy": 7.400482702255249,
      "epoch": 0.024223187437353827,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001445,
      "loss": 6.9731,
      "mean_token_accuracy": 0.0870087906718254,
      "num_tokens": 530631.0,
      "step": 290
    },
    {
      "entropy": 7.305219221115112,
      "epoch": 0.024640828600066823,
      "grad_norm": 1.328125,
      "learning_rate": 0.000147,
      "loss": 6.8415,
      "mean_token_accuracy": 0.10079794302582741,
      "num_tokens": 540386.0,
      "step": 295
    },
    {
      "entropy": 7.363999080657959,
      "epoch": 0.02505846976277982,
      "grad_norm": 1.5625,
      "learning_rate": 0.0001495,
      "loss": 6.9284,
      "mean_token_accuracy": 0.09086689502000808,
      "num_tokens": 550121.0,
      "step": 300
    },
    {
      "entropy": 7.154841661453247,
      "epoch": 0.025476110925492818,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000152,
      "loss": 6.7948,
      "mean_token_accuracy": 0.09512273818254471,
      "num_tokens": 559460.0,
      "step": 305
    },
    {
      "entropy": 7.282400894165039,
      "epoch": 0.025893752088205813,
      "grad_norm": 1.421875,
      "learning_rate": 0.00015450000000000001,
      "loss": 6.7419,
      "mean_token_accuracy": 0.10121037140488624,
      "num_tokens": 568392.0,
      "step": 310
    },
    {
      "entropy": 7.279255533218384,
      "epoch": 0.02631139325091881,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000157,
      "loss": 6.8451,
      "mean_token_accuracy": 0.09345789849758149,
      "num_tokens": 578248.0,
      "step": 315
    },
    {
      "entropy": 7.27221417427063,
      "epoch": 0.026729034413631808,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0001595,
      "loss": 6.9144,
      "mean_token_accuracy": 0.09365530461072921,
      "num_tokens": 588164.0,
      "step": 320
    },
    {
      "entropy": 7.197469997406006,
      "epoch": 0.027146675576344804,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000162,
      "loss": 6.8342,
      "mean_token_accuracy": 0.09261370748281479,
      "num_tokens": 598715.0,
      "step": 325
    },
    {
      "entropy": 7.274549579620361,
      "epoch": 0.027564316739057803,
      "grad_norm": 1.359375,
      "learning_rate": 0.00016450000000000001,
      "loss": 6.7664,
      "mean_token_accuracy": 0.09446420967578888,
      "num_tokens": 606939.0,
      "step": 330
    },
    {
      "entropy": 7.175267553329467,
      "epoch": 0.0279819579017708,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00016700000000000002,
      "loss": 6.8544,
      "mean_token_accuracy": 0.0952809251844883,
      "num_tokens": 615803.0,
      "step": 335
    },
    {
      "entropy": 7.135472536087036,
      "epoch": 0.028399599064483794,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00016950000000000003,
      "loss": 6.7134,
      "mean_token_accuracy": 0.0974463865160942,
      "num_tokens": 624730.0,
      "step": 340
    },
    {
      "entropy": 7.221632957458496,
      "epoch": 0.028817240227196794,
      "grad_norm": 1.484375,
      "learning_rate": 0.00017199999999999998,
      "loss": 6.8097,
      "mean_token_accuracy": 0.09591780230402946,
      "num_tokens": 634082.0,
      "step": 345
    },
    {
      "entropy": 7.172988319396973,
      "epoch": 0.02923488138990979,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00017449999999999999,
      "loss": 6.7906,
      "mean_token_accuracy": 0.09750410541892052,
      "num_tokens": 643367.0,
      "step": 350
    },
    {
      "entropy": 7.093995666503906,
      "epoch": 0.029652522552622785,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000177,
      "loss": 6.6148,
      "mean_token_accuracy": 0.10173115655779838,
      "num_tokens": 651861.0,
      "step": 355
    },
    {
      "entropy": 7.0821737289428714,
      "epoch": 0.030070163715335784,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0001795,
      "loss": 6.7571,
      "mean_token_accuracy": 0.0948550172150135,
      "num_tokens": 661064.0,
      "step": 360
    },
    {
      "entropy": 7.153480052947998,
      "epoch": 0.03048780487804878,
      "grad_norm": 1.34375,
      "learning_rate": 0.000182,
      "loss": 6.7307,
      "mean_token_accuracy": 0.10703616738319396,
      "num_tokens": 669299.0,
      "step": 365
    },
    {
      "entropy": 7.030750942230225,
      "epoch": 0.03090544604076178,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001845,
      "loss": 6.8063,
      "mean_token_accuracy": 0.09672270379960538,
      "num_tokens": 678631.0,
      "step": 370
    },
    {
      "entropy": 7.175200462341309,
      "epoch": 0.03132308720347477,
      "grad_norm": 1.359375,
      "learning_rate": 0.000187,
      "loss": 6.7998,
      "mean_token_accuracy": 0.09658469036221504,
      "num_tokens": 688511.0,
      "step": 375
    },
    {
      "entropy": 7.161194705963135,
      "epoch": 0.03174072836618777,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001895,
      "loss": 6.7375,
      "mean_token_accuracy": 0.09603499248623848,
      "num_tokens": 697956.0,
      "step": 380
    },
    {
      "entropy": 7.05234808921814,
      "epoch": 0.03215836952890077,
      "grad_norm": 1.34375,
      "learning_rate": 0.000192,
      "loss": 6.6413,
      "mean_token_accuracy": 0.10262225940823555,
      "num_tokens": 706779.0,
      "step": 385
    },
    {
      "entropy": 7.056643486022949,
      "epoch": 0.03257601069161377,
      "grad_norm": 1.1875,
      "learning_rate": 0.0001945,
      "loss": 6.6241,
      "mean_token_accuracy": 0.10087177231907844,
      "num_tokens": 715749.0,
      "step": 390
    },
    {
      "entropy": 6.9739141941070555,
      "epoch": 0.03299365185432676,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00019700000000000002,
      "loss": 6.6158,
      "mean_token_accuracy": 0.09956808760762215,
      "num_tokens": 724756.0,
      "step": 395
    },
    {
      "entropy": 6.9772045612335205,
      "epoch": 0.03341129301703976,
      "grad_norm": 1.375,
      "learning_rate": 0.00019950000000000002,
      "loss": 6.6336,
      "mean_token_accuracy": 0.09985274448990822,
      "num_tokens": 733288.0,
      "step": 400
    },
    {
      "entropy": 7.113308477401733,
      "epoch": 0.03382893417975276,
      "grad_norm": 1.4453125,
      "learning_rate": 0.000202,
      "loss": 6.7437,
      "mean_token_accuracy": 0.09781895428895951,
      "num_tokens": 741629.0,
      "step": 405
    },
    {
      "entropy": 7.011357355117798,
      "epoch": 0.03424657534246575,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00020449999999999998,
      "loss": 6.6754,
      "mean_token_accuracy": 0.10406573414802552,
      "num_tokens": 751336.0,
      "step": 410
    },
    {
      "entropy": 7.130186605453491,
      "epoch": 0.03466421650517875,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000207,
      "loss": 6.7513,
      "mean_token_accuracy": 0.09732639789581299,
      "num_tokens": 759725.0,
      "step": 415
    },
    {
      "entropy": 7.069876623153687,
      "epoch": 0.03508185766789175,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002095,
      "loss": 6.7657,
      "mean_token_accuracy": 0.09712367430329323,
      "num_tokens": 769752.0,
      "step": 420
    },
    {
      "entropy": 7.016468811035156,
      "epoch": 0.03549949883060474,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000212,
      "loss": 6.6428,
      "mean_token_accuracy": 0.10409499630331993,
      "num_tokens": 779124.0,
      "step": 425
    },
    {
      "entropy": 6.8530608177185055,
      "epoch": 0.03591713999331774,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002145,
      "loss": 6.602,
      "mean_token_accuracy": 0.10306296646595,
      "num_tokens": 788892.0,
      "step": 430
    },
    {
      "entropy": 6.865619564056397,
      "epoch": 0.03633478115603074,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00021700000000000002,
      "loss": 6.5647,
      "mean_token_accuracy": 0.10568981394171714,
      "num_tokens": 798265.0,
      "step": 435
    },
    {
      "entropy": 6.853310251235962,
      "epoch": 0.03675242231874373,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002195,
      "loss": 6.5224,
      "mean_token_accuracy": 0.10847331061959267,
      "num_tokens": 806896.0,
      "step": 440
    },
    {
      "entropy": 7.006761074066162,
      "epoch": 0.03717006348145673,
      "grad_norm": 1.515625,
      "learning_rate": 0.000222,
      "loss": 6.5879,
      "mean_token_accuracy": 0.10766485631465912,
      "num_tokens": 815540.0,
      "step": 445
    },
    {
      "entropy": 6.854085683822632,
      "epoch": 0.03758770464416973,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002245,
      "loss": 6.6495,
      "mean_token_accuracy": 0.10323781818151474,
      "num_tokens": 824555.0,
      "step": 450
    },
    {
      "entropy": 7.022783946990967,
      "epoch": 0.03800534580688272,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00022700000000000002,
      "loss": 6.6988,
      "mean_token_accuracy": 0.10127760544419288,
      "num_tokens": 833414.0,
      "step": 455
    },
    {
      "entropy": 6.811462020874023,
      "epoch": 0.03842298696959572,
      "grad_norm": 1.203125,
      "learning_rate": 0.00022950000000000002,
      "loss": 6.5436,
      "mean_token_accuracy": 0.11053832620382309,
      "num_tokens": 842644.0,
      "step": 460
    },
    {
      "entropy": 6.894090795516968,
      "epoch": 0.03884062813230872,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00023200000000000003,
      "loss": 6.5682,
      "mean_token_accuracy": 0.10382608100771903,
      "num_tokens": 851786.0,
      "step": 465
    },
    {
      "entropy": 6.843556022644043,
      "epoch": 0.03925826929502172,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00023449999999999998,
      "loss": 6.6229,
      "mean_token_accuracy": 0.10601722747087479,
      "num_tokens": 861508.0,
      "step": 470
    },
    {
      "entropy": 6.869661664962768,
      "epoch": 0.03967591045773471,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000237,
      "loss": 6.4919,
      "mean_token_accuracy": 0.10929096192121505,
      "num_tokens": 870433.0,
      "step": 475
    },
    {
      "entropy": 6.916244745254517,
      "epoch": 0.04009355162044771,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002395,
      "loss": 6.6124,
      "mean_token_accuracy": 0.1070750541985035,
      "num_tokens": 879059.0,
      "step": 480
    },
    {
      "entropy": 6.885466527938843,
      "epoch": 0.04051119278316071,
      "grad_norm": 1.28125,
      "learning_rate": 0.000242,
      "loss": 6.6607,
      "mean_token_accuracy": 0.10065596923232079,
      "num_tokens": 888009.0,
      "step": 485
    },
    {
      "entropy": 6.7803881645202635,
      "epoch": 0.040928833945873704,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002445,
      "loss": 6.529,
      "mean_token_accuracy": 0.10448121652007103,
      "num_tokens": 897335.0,
      "step": 490
    },
    {
      "entropy": 6.953722333908081,
      "epoch": 0.0413464751085867,
      "grad_norm": 1.25,
      "learning_rate": 0.000247,
      "loss": 6.5241,
      "mean_token_accuracy": 0.10960079357028008,
      "num_tokens": 905802.0,
      "step": 495
    },
    {
      "entropy": 6.839751958847046,
      "epoch": 0.0417641162712997,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002495,
      "loss": 6.5288,
      "mean_token_accuracy": 0.10939614027738571,
      "num_tokens": 914836.0,
      "step": 500
    },
    {
      "entropy": 6.722894287109375,
      "epoch": 0.042181757434012694,
      "grad_norm": 1.296875,
      "learning_rate": 0.000252,
      "loss": 6.4752,
      "mean_token_accuracy": 0.11474435329437256,
      "num_tokens": 924254.0,
      "step": 505
    },
    {
      "entropy": 6.799596691131592,
      "epoch": 0.04259939859672569,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002545,
      "loss": 6.5715,
      "mean_token_accuracy": 0.10900455489754676,
      "num_tokens": 933841.0,
      "step": 510
    },
    {
      "entropy": 6.8436116695404055,
      "epoch": 0.04301703975943869,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000257,
      "loss": 6.6245,
      "mean_token_accuracy": 0.10766134932637214,
      "num_tokens": 943304.0,
      "step": 515
    },
    {
      "entropy": 6.756516551971435,
      "epoch": 0.043434680922151685,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002595,
      "loss": 6.4616,
      "mean_token_accuracy": 0.11000765487551689,
      "num_tokens": 952915.0,
      "step": 520
    },
    {
      "entropy": 6.807320642471313,
      "epoch": 0.043852322084864684,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000262,
      "loss": 6.5169,
      "mean_token_accuracy": 0.11124896109104157,
      "num_tokens": 962752.0,
      "step": 525
    },
    {
      "entropy": 6.805405187606811,
      "epoch": 0.04426996324757768,
      "grad_norm": 1.3125,
      "learning_rate": 0.00026450000000000003,
      "loss": 6.5271,
      "mean_token_accuracy": 0.09916252493858338,
      "num_tokens": 971240.0,
      "step": 530
    },
    {
      "entropy": 6.747521448135376,
      "epoch": 0.044687604410290675,
      "grad_norm": 1.3125,
      "learning_rate": 0.00026700000000000004,
      "loss": 6.5801,
      "mean_token_accuracy": 0.10155974254012108,
      "num_tokens": 981391.0,
      "step": 535
    },
    {
      "entropy": 6.818302583694458,
      "epoch": 0.045105245573003674,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00026950000000000005,
      "loss": 6.5666,
      "mean_token_accuracy": 0.1143165647983551,
      "num_tokens": 990551.0,
      "step": 540
    },
    {
      "entropy": 6.7388129234313965,
      "epoch": 0.045522886735716674,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00027200000000000005,
      "loss": 6.4962,
      "mean_token_accuracy": 0.11189664974808693,
      "num_tokens": 999593.0,
      "step": 545
    },
    {
      "entropy": 6.785442161560058,
      "epoch": 0.045940527898429666,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002745,
      "loss": 6.5789,
      "mean_token_accuracy": 0.10544709712266923,
      "num_tokens": 1008785.0,
      "step": 550
    },
    {
      "entropy": 6.757883024215698,
      "epoch": 0.046358169061142665,
      "grad_norm": 1.3125,
      "learning_rate": 0.000277,
      "loss": 6.5604,
      "mean_token_accuracy": 0.10769657045602798,
      "num_tokens": 1016815.0,
      "step": 555
    },
    {
      "entropy": 6.758020496368408,
      "epoch": 0.046775810223855664,
      "grad_norm": 1.25,
      "learning_rate": 0.0002795,
      "loss": 6.4344,
      "mean_token_accuracy": 0.11475159376859664,
      "num_tokens": 1026182.0,
      "step": 560
    },
    {
      "entropy": 6.683844518661499,
      "epoch": 0.04719345138656866,
      "grad_norm": 1.53125,
      "learning_rate": 0.00028199999999999997,
      "loss": 6.3739,
      "mean_token_accuracy": 0.11326156556606293,
      "num_tokens": 1034032.0,
      "step": 565
    },
    {
      "entropy": 6.730586194992066,
      "epoch": 0.047611092549281656,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002845,
      "loss": 6.5731,
      "mean_token_accuracy": 0.10969387739896774,
      "num_tokens": 1043493.0,
      "step": 570
    },
    {
      "entropy": 6.746386432647705,
      "epoch": 0.048028733711994655,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000287,
      "loss": 6.4697,
      "mean_token_accuracy": 0.11479877978563309,
      "num_tokens": 1052989.0,
      "step": 575
    },
    {
      "entropy": 6.649836444854737,
      "epoch": 0.048446374874707654,
      "grad_norm": 1.25,
      "learning_rate": 0.0002895,
      "loss": 6.4517,
      "mean_token_accuracy": 0.11029219925403595,
      "num_tokens": 1061803.0,
      "step": 580
    },
    {
      "entropy": 6.773694610595703,
      "epoch": 0.048864016037420646,
      "grad_norm": 1.09375,
      "learning_rate": 0.000292,
      "loss": 6.5561,
      "mean_token_accuracy": 0.10948128178715706,
      "num_tokens": 1071227.0,
      "step": 585
    },
    {
      "entropy": 6.727704048156738,
      "epoch": 0.049281657200133645,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002945,
      "loss": 6.5617,
      "mean_token_accuracy": 0.11164849177002907,
      "num_tokens": 1080082.0,
      "step": 590
    },
    {
      "entropy": 6.657727670669556,
      "epoch": 0.049699298362846644,
      "grad_norm": 1.28125,
      "learning_rate": 0.000297,
      "loss": 6.3975,
      "mean_token_accuracy": 0.11562075838446617,
      "num_tokens": 1089750.0,
      "step": 595
    },
    {
      "entropy": 6.567789506912232,
      "epoch": 0.05011693952555964,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002995,
      "loss": 6.3956,
      "mean_token_accuracy": 0.11055527701973915,
      "num_tokens": 1098691.0,
      "step": 600
    },
    {
      "entropy": 6.651774883270264,
      "epoch": 0.050534580688272636,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000302,
      "loss": 6.4191,
      "mean_token_accuracy": 0.11506491899490356,
      "num_tokens": 1107602.0,
      "step": 605
    },
    {
      "entropy": 6.595056581497192,
      "epoch": 0.050952221850985635,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003045,
      "loss": 6.3422,
      "mean_token_accuracy": 0.11736676096916199,
      "num_tokens": 1116426.0,
      "step": 610
    },
    {
      "entropy": 6.746117305755615,
      "epoch": 0.05136986301369863,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000307,
      "loss": 6.5262,
      "mean_token_accuracy": 0.11227416917681694,
      "num_tokens": 1126221.0,
      "step": 615
    },
    {
      "entropy": 6.536876583099366,
      "epoch": 0.051787504176411626,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003095,
      "loss": 6.401,
      "mean_token_accuracy": 0.11846380457282066,
      "num_tokens": 1135662.0,
      "step": 620
    },
    {
      "entropy": 6.675752067565918,
      "epoch": 0.052205145339124626,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000312,
      "loss": 6.5017,
      "mean_token_accuracy": 0.11206841841340065,
      "num_tokens": 1143963.0,
      "step": 625
    },
    {
      "entropy": 6.615093946456909,
      "epoch": 0.05262278650183762,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003145,
      "loss": 6.3808,
      "mean_token_accuracy": 0.12234973758459092,
      "num_tokens": 1152470.0,
      "step": 630
    },
    {
      "entropy": 6.61521053314209,
      "epoch": 0.05304042766455062,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000317,
      "loss": 6.3481,
      "mean_token_accuracy": 0.11505335196852684,
      "num_tokens": 1161257.0,
      "step": 635
    },
    {
      "entropy": 6.679902791976929,
      "epoch": 0.053458068827263616,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003195,
      "loss": 6.4666,
      "mean_token_accuracy": 0.11135088875889779,
      "num_tokens": 1170969.0,
      "step": 640
    },
    {
      "entropy": 6.6380119800567625,
      "epoch": 0.053875709989976615,
      "grad_norm": 1.078125,
      "learning_rate": 0.000322,
      "loss": 6.6069,
      "mean_token_accuracy": 0.11080317497253418,
      "num_tokens": 1181127.0,
      "step": 645
    },
    {
      "entropy": 6.6212114810943605,
      "epoch": 0.05429335115268961,
      "grad_norm": 1.3125,
      "learning_rate": 0.00032450000000000003,
      "loss": 6.4079,
      "mean_token_accuracy": 0.11520254909992218,
      "num_tokens": 1190260.0,
      "step": 650
    },
    {
      "entropy": 6.626252603530884,
      "epoch": 0.05471099231540261,
      "grad_norm": 1.203125,
      "learning_rate": 0.00032700000000000003,
      "loss": 6.483,
      "mean_token_accuracy": 0.11253710016608238,
      "num_tokens": 1200013.0,
      "step": 655
    },
    {
      "entropy": 6.691232061386108,
      "epoch": 0.055128633478115606,
      "grad_norm": 1.5625,
      "learning_rate": 0.00032950000000000004,
      "loss": 6.42,
      "mean_token_accuracy": 0.1188363939523697,
      "num_tokens": 1209239.0,
      "step": 660
    },
    {
      "entropy": 6.489693546295166,
      "epoch": 0.0555462746408286,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00033200000000000005,
      "loss": 6.3535,
      "mean_token_accuracy": 0.11323748677968978,
      "num_tokens": 1218552.0,
      "step": 665
    },
    {
      "entropy": 6.521316194534302,
      "epoch": 0.0559639158035416,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00033450000000000005,
      "loss": 6.4197,
      "mean_token_accuracy": 0.1132273718714714,
      "num_tokens": 1227632.0,
      "step": 670
    },
    {
      "entropy": 6.628327083587647,
      "epoch": 0.056381556966254597,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000337,
      "loss": 6.4666,
      "mean_token_accuracy": 0.11458859890699387,
      "num_tokens": 1236433.0,
      "step": 675
    },
    {
      "entropy": 6.535047388076782,
      "epoch": 0.05679919812896759,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003395,
      "loss": 6.4887,
      "mean_token_accuracy": 0.10250034034252167,
      "num_tokens": 1245700.0,
      "step": 680
    },
    {
      "entropy": 6.58137092590332,
      "epoch": 0.05721683929168059,
      "grad_norm": 1.1875,
      "learning_rate": 0.000342,
      "loss": 6.3208,
      "mean_token_accuracy": 0.11790363714098931,
      "num_tokens": 1254206.0,
      "step": 685
    },
    {
      "entropy": 6.546811771392822,
      "epoch": 0.05763448045439359,
      "grad_norm": 1.171875,
      "learning_rate": 0.00034449999999999997,
      "loss": 6.3787,
      "mean_token_accuracy": 0.11770905256271362,
      "num_tokens": 1263302.0,
      "step": 690
    },
    {
      "entropy": 6.6236083984375,
      "epoch": 0.05805212161710658,
      "grad_norm": 1.15625,
      "learning_rate": 0.000347,
      "loss": 6.4224,
      "mean_token_accuracy": 0.11309175118803978,
      "num_tokens": 1272998.0,
      "step": 695
    },
    {
      "entropy": 6.480911111831665,
      "epoch": 0.05846976277981958,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003495,
      "loss": 6.396,
      "mean_token_accuracy": 0.11966503635048867,
      "num_tokens": 1281955.0,
      "step": 700
    },
    {
      "entropy": 6.559855842590332,
      "epoch": 0.05888740394253258,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000352,
      "loss": 6.4235,
      "mean_token_accuracy": 0.11396420970559121,
      "num_tokens": 1291219.0,
      "step": 705
    },
    {
      "entropy": 6.5371620655059814,
      "epoch": 0.05930504510524557,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003545,
      "loss": 6.4288,
      "mean_token_accuracy": 0.11189143359661102,
      "num_tokens": 1300963.0,
      "step": 710
    },
    {
      "entropy": 6.616220235824585,
      "epoch": 0.05972268626795857,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000357,
      "loss": 6.4596,
      "mean_token_accuracy": 0.11059518307447433,
      "num_tokens": 1310112.0,
      "step": 715
    },
    {
      "entropy": 6.549956035614014,
      "epoch": 0.06014032743067157,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003595,
      "loss": 6.3884,
      "mean_token_accuracy": 0.12019749581813813,
      "num_tokens": 1319309.0,
      "step": 720
    },
    {
      "entropy": 6.4225273609161375,
      "epoch": 0.06055796859338457,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000362,
      "loss": 6.3737,
      "mean_token_accuracy": 0.11447888761758804,
      "num_tokens": 1329049.0,
      "step": 725
    },
    {
      "entropy": 6.456729459762573,
      "epoch": 0.06097560975609756,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003645,
      "loss": 6.2122,
      "mean_token_accuracy": 0.11689538508653641,
      "num_tokens": 1337770.0,
      "step": 730
    },
    {
      "entropy": 6.3992725849151615,
      "epoch": 0.06139325091881056,
      "grad_norm": 1.296875,
      "learning_rate": 0.000367,
      "loss": 6.3188,
      "mean_token_accuracy": 0.11989094093441963,
      "num_tokens": 1346705.0,
      "step": 735
    },
    {
      "entropy": 6.454647445678711,
      "epoch": 0.06181089208152356,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003695,
      "loss": 6.3721,
      "mean_token_accuracy": 0.11653787791728973,
      "num_tokens": 1354860.0,
      "step": 740
    },
    {
      "entropy": 6.486315011978149,
      "epoch": 0.06222853324423655,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000372,
      "loss": 6.3431,
      "mean_token_accuracy": 0.11571619734168052,
      "num_tokens": 1364140.0,
      "step": 745
    },
    {
      "entropy": 6.577937173843384,
      "epoch": 0.06264617440694954,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003745,
      "loss": 6.474,
      "mean_token_accuracy": 0.10803929641842842,
      "num_tokens": 1373645.0,
      "step": 750
    },
    {
      "entropy": 6.493966341018677,
      "epoch": 0.06306381556966255,
      "grad_norm": 1.109375,
      "learning_rate": 0.000377,
      "loss": 6.3496,
      "mean_token_accuracy": 0.12465299069881439,
      "num_tokens": 1383291.0,
      "step": 755
    },
    {
      "entropy": 6.395097112655639,
      "epoch": 0.06348145673237554,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003795,
      "loss": 6.2609,
      "mean_token_accuracy": 0.12159460708498955,
      "num_tokens": 1391172.0,
      "step": 760
    },
    {
      "entropy": 6.490391969680786,
      "epoch": 0.06389909789508853,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000382,
      "loss": 6.2816,
      "mean_token_accuracy": 0.11841546595096589,
      "num_tokens": 1400864.0,
      "step": 765
    },
    {
      "entropy": 6.427747774124145,
      "epoch": 0.06431673905780154,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003845,
      "loss": 6.3416,
      "mean_token_accuracy": 0.12262844294309616,
      "num_tokens": 1410570.0,
      "step": 770
    },
    {
      "entropy": 6.404248476028442,
      "epoch": 0.06473438022051453,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00038700000000000003,
      "loss": 6.3088,
      "mean_token_accuracy": 0.11854264065623284,
      "num_tokens": 1419536.0,
      "step": 775
    },
    {
      "entropy": 6.388956737518311,
      "epoch": 0.06515202138322754,
      "grad_norm": 1.1875,
      "learning_rate": 0.00038950000000000003,
      "loss": 6.2301,
      "mean_token_accuracy": 0.12381713986396789,
      "num_tokens": 1428000.0,
      "step": 780
    },
    {
      "entropy": 6.417420148849487,
      "epoch": 0.06556966254594053,
      "grad_norm": 1.28125,
      "learning_rate": 0.00039200000000000004,
      "loss": 6.4262,
      "mean_token_accuracy": 0.11879310831427574,
      "num_tokens": 1437789.0,
      "step": 785
    },
    {
      "entropy": 6.386681938171387,
      "epoch": 0.06598730370865352,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039450000000000005,
      "loss": 6.2391,
      "mean_token_accuracy": 0.12141718417406082,
      "num_tokens": 1446848.0,
      "step": 790
    },
    {
      "entropy": 6.396657800674438,
      "epoch": 0.06640494487136653,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039700000000000005,
      "loss": 6.2814,
      "mean_token_accuracy": 0.1192056655883789,
      "num_tokens": 1455749.0,
      "step": 795
    },
    {
      "entropy": 6.357221508026123,
      "epoch": 0.06682258603407952,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003995,
      "loss": 6.2013,
      "mean_token_accuracy": 0.12639149650931358,
      "num_tokens": 1464556.0,
      "step": 800
    },
    {
      "entropy": 6.398760557174683,
      "epoch": 0.06724022719679251,
      "grad_norm": 1.25,
      "learning_rate": 0.000402,
      "loss": 6.3614,
      "mean_token_accuracy": 0.11781241819262504,
      "num_tokens": 1473617.0,
      "step": 805
    },
    {
      "entropy": 6.494735813140869,
      "epoch": 0.06765786835950552,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004045,
      "loss": 6.3559,
      "mean_token_accuracy": 0.11977834030985832,
      "num_tokens": 1483006.0,
      "step": 810
    },
    {
      "entropy": 6.457280921936035,
      "epoch": 0.06807550952221851,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00040699999999999997,
      "loss": 6.3836,
      "mean_token_accuracy": 0.11731792315840721,
      "num_tokens": 1493009.0,
      "step": 815
    },
    {
      "entropy": 6.363678503036499,
      "epoch": 0.0684931506849315,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004095,
      "loss": 6.2556,
      "mean_token_accuracy": 0.12392903044819832,
      "num_tokens": 1501261.0,
      "step": 820
    },
    {
      "entropy": 6.3693873405456545,
      "epoch": 0.06891079184764451,
      "grad_norm": 1.03125,
      "learning_rate": 0.000412,
      "loss": 6.3709,
      "mean_token_accuracy": 0.11665059328079223,
      "num_tokens": 1511623.0,
      "step": 825
    },
    {
      "entropy": 6.428213024139405,
      "epoch": 0.0693284330103575,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004145,
      "loss": 6.3659,
      "mean_token_accuracy": 0.11537070199847221,
      "num_tokens": 1521741.0,
      "step": 830
    },
    {
      "entropy": 6.409314584732056,
      "epoch": 0.0697460741730705,
      "grad_norm": 1.03125,
      "learning_rate": 0.000417,
      "loss": 6.3035,
      "mean_token_accuracy": 0.12089102566242219,
      "num_tokens": 1530988.0,
      "step": 835
    },
    {
      "entropy": 6.339814519882202,
      "epoch": 0.0701637153357835,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004195,
      "loss": 6.1285,
      "mean_token_accuracy": 0.12743735313415527,
      "num_tokens": 1539415.0,
      "step": 840
    },
    {
      "entropy": 6.378707456588745,
      "epoch": 0.07058135649849649,
      "grad_norm": 1.234375,
      "learning_rate": 0.000422,
      "loss": 6.3338,
      "mean_token_accuracy": 0.11762123182415962,
      "num_tokens": 1547969.0,
      "step": 845
    },
    {
      "entropy": 6.275123023986817,
      "epoch": 0.07099899766120948,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004245,
      "loss": 6.193,
      "mean_token_accuracy": 0.12221964225172996,
      "num_tokens": 1557526.0,
      "step": 850
    },
    {
      "entropy": 6.299358463287353,
      "epoch": 0.07141663882392249,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000427,
      "loss": 6.1699,
      "mean_token_accuracy": 0.12564374953508378,
      "num_tokens": 1565982.0,
      "step": 855
    },
    {
      "entropy": 6.304647016525268,
      "epoch": 0.07183427998663548,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004295,
      "loss": 6.2293,
      "mean_token_accuracy": 0.12849857807159423,
      "num_tokens": 1574492.0,
      "step": 860
    },
    {
      "entropy": 6.269746017456055,
      "epoch": 0.07225192114934847,
      "grad_norm": 1.171875,
      "learning_rate": 0.000432,
      "loss": 6.2051,
      "mean_token_accuracy": 0.12123230844736099,
      "num_tokens": 1583726.0,
      "step": 865
    },
    {
      "entropy": 6.284060335159301,
      "epoch": 0.07266956231206148,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004345,
      "loss": 6.2154,
      "mean_token_accuracy": 0.11842950582504272,
      "num_tokens": 1593780.0,
      "step": 870
    },
    {
      "entropy": 6.351301908493042,
      "epoch": 0.07308720347477447,
      "grad_norm": 1.171875,
      "learning_rate": 0.000437,
      "loss": 6.2104,
      "mean_token_accuracy": 0.12493377402424813,
      "num_tokens": 1602746.0,
      "step": 875
    },
    {
      "entropy": 6.353745460510254,
      "epoch": 0.07350484463748747,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004395,
      "loss": 6.2485,
      "mean_token_accuracy": 0.12405433133244514,
      "num_tokens": 1611767.0,
      "step": 880
    },
    {
      "entropy": 6.375559568405151,
      "epoch": 0.07392248580020047,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000442,
      "loss": 6.3327,
      "mean_token_accuracy": 0.12374300360679627,
      "num_tokens": 1621568.0,
      "step": 885
    },
    {
      "entropy": 6.310373783111572,
      "epoch": 0.07434012696291346,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004445,
      "loss": 6.2722,
      "mean_token_accuracy": 0.12864986136555673,
      "num_tokens": 1630650.0,
      "step": 890
    },
    {
      "entropy": 6.326134872436524,
      "epoch": 0.07475776812562646,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000447,
      "loss": 6.2808,
      "mean_token_accuracy": 0.11709838211536408,
      "num_tokens": 1640482.0,
      "step": 895
    },
    {
      "entropy": 6.344818496704102,
      "epoch": 0.07517540928833946,
      "grad_norm": 1.25,
      "learning_rate": 0.00044950000000000003,
      "loss": 6.3133,
      "mean_token_accuracy": 0.12154146954417229,
      "num_tokens": 1649344.0,
      "step": 900
    },
    {
      "entropy": 6.331745386123657,
      "epoch": 0.07559305045105245,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045200000000000004,
      "loss": 6.2463,
      "mean_token_accuracy": 0.12590100318193437,
      "num_tokens": 1658962.0,
      "step": 905
    },
    {
      "entropy": 6.229383707046509,
      "epoch": 0.07601069161376545,
      "grad_norm": 1.125,
      "learning_rate": 0.00045450000000000004,
      "loss": 6.2623,
      "mean_token_accuracy": 0.12040851265192032,
      "num_tokens": 1668615.0,
      "step": 910
    },
    {
      "entropy": 6.35636830329895,
      "epoch": 0.07642833277647845,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045700000000000005,
      "loss": 6.1982,
      "mean_token_accuracy": 0.12516034319996833,
      "num_tokens": 1679282.0,
      "step": 915
    },
    {
      "entropy": 6.20735993385315,
      "epoch": 0.07684597393919144,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045950000000000006,
      "loss": 6.2958,
      "mean_token_accuracy": 0.1260470613837242,
      "num_tokens": 1688362.0,
      "step": 920
    },
    {
      "entropy": 6.310071229934692,
      "epoch": 0.07726361510190444,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000462,
      "loss": 6.1767,
      "mean_token_accuracy": 0.12261404544115066,
      "num_tokens": 1697593.0,
      "step": 925
    },
    {
      "entropy": 6.316716861724854,
      "epoch": 0.07768125626461744,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004645,
      "loss": 6.2492,
      "mean_token_accuracy": 0.12079507783055306,
      "num_tokens": 1706961.0,
      "step": 930
    },
    {
      "entropy": 6.359928178787231,
      "epoch": 0.07809889742733044,
      "grad_norm": 1.15625,
      "learning_rate": 0.000467,
      "loss": 6.2909,
      "mean_token_accuracy": 0.1333912968635559,
      "num_tokens": 1716222.0,
      "step": 935
    },
    {
      "entropy": 6.185659694671631,
      "epoch": 0.07851653859004344,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004695,
      "loss": 6.1649,
      "mean_token_accuracy": 0.11959992647171021,
      "num_tokens": 1726750.0,
      "step": 940
    },
    {
      "entropy": 6.369044351577759,
      "epoch": 0.07893417975275643,
      "grad_norm": 1.0625,
      "learning_rate": 0.000472,
      "loss": 6.2557,
      "mean_token_accuracy": 0.1164586953818798,
      "num_tokens": 1736262.0,
      "step": 945
    },
    {
      "entropy": 6.249402666091919,
      "epoch": 0.07935182091546943,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004745,
      "loss": 6.2714,
      "mean_token_accuracy": 0.12722517997026445,
      "num_tokens": 1745555.0,
      "step": 950
    },
    {
      "entropy": 6.137383937835693,
      "epoch": 0.07976946207818243,
      "grad_norm": 1.234375,
      "learning_rate": 0.000477,
      "loss": 6.1563,
      "mean_token_accuracy": 0.12104023173451424,
      "num_tokens": 1754107.0,
      "step": 955
    },
    {
      "entropy": 6.396907472610474,
      "epoch": 0.08018710324089542,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004795,
      "loss": 6.3493,
      "mean_token_accuracy": 0.1176336333155632,
      "num_tokens": 1762982.0,
      "step": 960
    },
    {
      "entropy": 6.245203161239624,
      "epoch": 0.08060474440360842,
      "grad_norm": 1.140625,
      "learning_rate": 0.000482,
      "loss": 6.2228,
      "mean_token_accuracy": 0.12653560936450958,
      "num_tokens": 1772545.0,
      "step": 965
    },
    {
      "entropy": 6.3770890712738035,
      "epoch": 0.08102238556632142,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004845,
      "loss": 6.3507,
      "mean_token_accuracy": 0.11759537160396576,
      "num_tokens": 1782052.0,
      "step": 970
    },
    {
      "entropy": 6.178989744186401,
      "epoch": 0.08144002672903441,
      "grad_norm": 1.0625,
      "learning_rate": 0.000487,
      "loss": 6.1567,
      "mean_token_accuracy": 0.12350800633430481,
      "num_tokens": 1791682.0,
      "step": 975
    },
    {
      "entropy": 6.251344394683838,
      "epoch": 0.08185766789174741,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004895,
      "loss": 6.1646,
      "mean_token_accuracy": 0.12833068072795867,
      "num_tokens": 1800526.0,
      "step": 980
    },
    {
      "entropy": 6.298021268844605,
      "epoch": 0.08227530905446041,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000492,
      "loss": 6.2999,
      "mean_token_accuracy": 0.1231074184179306,
      "num_tokens": 1810062.0,
      "step": 985
    },
    {
      "entropy": 6.250222015380859,
      "epoch": 0.0826929502171734,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004945,
      "loss": 6.2841,
      "mean_token_accuracy": 0.12629790231585503,
      "num_tokens": 1818909.0,
      "step": 990
    },
    {
      "entropy": 6.297880029678344,
      "epoch": 0.0831105913798864,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000497,
      "loss": 6.2327,
      "mean_token_accuracy": 0.13256481513381005,
      "num_tokens": 1827978.0,
      "step": 995
    },
    {
      "entropy": 6.1916309833526615,
      "epoch": 0.0835282325425994,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004995,
      "loss": 6.2013,
      "mean_token_accuracy": 0.12940397188067437,
      "num_tokens": 1837955.0,
      "step": 1000
    },
    {
      "entropy": 6.246883821487427,
      "epoch": 0.0839458737053124,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999999987395534,
      "loss": 6.2853,
      "mean_token_accuracy": 0.12494878992438316,
      "num_tokens": 1847699.0,
      "step": 1005
    },
    {
      "entropy": 6.375528335571289,
      "epoch": 0.08436351486802539,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999999936189888,
      "loss": 6.218,
      "mean_token_accuracy": 0.12546755075454713,
      "num_tokens": 1856490.0,
      "step": 1010
    },
    {
      "entropy": 6.195798301696778,
      "epoch": 0.0847811560307384,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999999845595285,
      "loss": 6.2219,
      "mean_token_accuracy": 0.1272422730922699,
      "num_tokens": 1865644.0,
      "step": 1015
    },
    {
      "entropy": 6.2077044486999515,
      "epoch": 0.08519879719345139,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004999999715611727,
      "loss": 6.1602,
      "mean_token_accuracy": 0.12565038353204727,
      "num_tokens": 1874450.0,
      "step": 1020
    },
    {
      "entropy": 6.16418948173523,
      "epoch": 0.08561643835616438,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999999546239215,
      "loss": 6.1804,
      "mean_token_accuracy": 0.12442194074392318,
      "num_tokens": 1883259.0,
      "step": 1025
    },
    {
      "entropy": 6.259844875335693,
      "epoch": 0.08603407951887738,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999999337477753,
      "loss": 6.2624,
      "mean_token_accuracy": 0.12271653115749359,
      "num_tokens": 1891897.0,
      "step": 1030
    },
    {
      "entropy": 6.2629540920257565,
      "epoch": 0.08645172068159038,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999999089327344,
      "loss": 6.2408,
      "mean_token_accuracy": 0.12325521185994148,
      "num_tokens": 1901011.0,
      "step": 1035
    },
    {
      "entropy": 6.100732040405274,
      "epoch": 0.08686936184430337,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004999998801787993,
      "loss": 6.1432,
      "mean_token_accuracy": 0.13062310665845872,
      "num_tokens": 1910132.0,
      "step": 1040
    },
    {
      "entropy": 6.287595129013061,
      "epoch": 0.08728700300701638,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999998474859704,
      "loss": 6.259,
      "mean_token_accuracy": 0.12846295088529586,
      "num_tokens": 1919938.0,
      "step": 1045
    },
    {
      "entropy": 6.134413957595825,
      "epoch": 0.08770464416972937,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999998108542483,
      "loss": 6.1267,
      "mean_token_accuracy": 0.12628429979085923,
      "num_tokens": 1928199.0,
      "step": 1050
    },
    {
      "entropy": 6.208824634552002,
      "epoch": 0.08812228533244236,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004999997702836339,
      "loss": 6.1162,
      "mean_token_accuracy": 0.12950822934508324,
      "num_tokens": 1937133.0,
      "step": 1055
    },
    {
      "entropy": 6.2759356021881105,
      "epoch": 0.08853992649515537,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999997257741274,
      "loss": 6.2783,
      "mean_token_accuracy": 0.1283426858484745,
      "num_tokens": 1946580.0,
      "step": 1060
    },
    {
      "entropy": 6.123747396469116,
      "epoch": 0.08895756765786836,
      "grad_norm": 1.09375,
      "learning_rate": 0.00049999967732573,
      "loss": 6.2076,
      "mean_token_accuracy": 0.12349388226866723,
      "num_tokens": 1955413.0,
      "step": 1065
    },
    {
      "entropy": 6.401907300949096,
      "epoch": 0.08937520882058135,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999996249384425,
      "loss": 6.2811,
      "mean_token_accuracy": 0.12440942004323005,
      "num_tokens": 1964716.0,
      "step": 1070
    },
    {
      "entropy": 6.09635558128357,
      "epoch": 0.08979284998329436,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999995686122657,
      "loss": 6.1292,
      "mean_token_accuracy": 0.1341006152331829,
      "num_tokens": 1974081.0,
      "step": 1075
    },
    {
      "entropy": 6.034217596054077,
      "epoch": 0.09021049114600735,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999995083472006,
      "loss": 6.0911,
      "mean_token_accuracy": 0.12835043892264367,
      "num_tokens": 1983035.0,
      "step": 1080
    },
    {
      "entropy": 6.26309871673584,
      "epoch": 0.09062813230872034,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999994441432481,
      "loss": 6.1167,
      "mean_token_accuracy": 0.13723048493266105,
      "num_tokens": 1992475.0,
      "step": 1085
    },
    {
      "entropy": 6.088156604766846,
      "epoch": 0.09104577347143335,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999993760004097,
      "loss": 6.1961,
      "mean_token_accuracy": 0.12458470240235328,
      "num_tokens": 2002729.0,
      "step": 1090
    },
    {
      "entropy": 6.078368520736694,
      "epoch": 0.09146341463414634,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999993039186864,
      "loss": 6.0401,
      "mean_token_accuracy": 0.1297457695007324,
      "num_tokens": 2011319.0,
      "step": 1095
    },
    {
      "entropy": 6.207320833206177,
      "epoch": 0.09188105579685933,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999992278980794,
      "loss": 6.0821,
      "mean_token_accuracy": 0.12979096546769142,
      "num_tokens": 2020456.0,
      "step": 1100
    },
    {
      "entropy": 6.020119953155517,
      "epoch": 0.09229869695957234,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004999991479385901,
      "loss": 6.1668,
      "mean_token_accuracy": 0.13272762671113014,
      "num_tokens": 2029417.0,
      "step": 1105
    },
    {
      "entropy": 6.307196569442749,
      "epoch": 0.09271633812228533,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999990640402198,
      "loss": 6.2571,
      "mean_token_accuracy": 0.12967613935470582,
      "num_tokens": 2039290.0,
      "step": 1110
    },
    {
      "entropy": 6.152990007400513,
      "epoch": 0.09313397928499834,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999989762029701,
      "loss": 6.0688,
      "mean_token_accuracy": 0.13426292017102243,
      "num_tokens": 2048485.0,
      "step": 1115
    },
    {
      "entropy": 5.983977270126343,
      "epoch": 0.09355162044771133,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999988844268425,
      "loss": 6.0518,
      "mean_token_accuracy": 0.13763293251395226,
      "num_tokens": 2057131.0,
      "step": 1120
    },
    {
      "entropy": 6.102506446838379,
      "epoch": 0.09396926161042432,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999987887118387,
      "loss": 6.0427,
      "mean_token_accuracy": 0.12984845265746117,
      "num_tokens": 2066295.0,
      "step": 1125
    },
    {
      "entropy": 6.096871328353882,
      "epoch": 0.09438690277313733,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999986890579602,
      "loss": 6.0481,
      "mean_token_accuracy": 0.13764009922742843,
      "num_tokens": 2075656.0,
      "step": 1130
    },
    {
      "entropy": 6.084485101699829,
      "epoch": 0.09480454393585032,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004999985854652087,
      "loss": 6.2038,
      "mean_token_accuracy": 0.12579603642225265,
      "num_tokens": 2086046.0,
      "step": 1135
    },
    {
      "entropy": 6.256638288497925,
      "epoch": 0.09522218509856331,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999984779335862,
      "loss": 6.2003,
      "mean_token_accuracy": 0.12873548045754432,
      "num_tokens": 2094808.0,
      "step": 1140
    },
    {
      "entropy": 6.063794231414795,
      "epoch": 0.09563982626127632,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999983664630945,
      "loss": 6.1302,
      "mean_token_accuracy": 0.13587065562605857,
      "num_tokens": 2104155.0,
      "step": 1145
    },
    {
      "entropy": 6.138120031356811,
      "epoch": 0.09605746742398931,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999982510537357,
      "loss": 6.0142,
      "mean_token_accuracy": 0.13489947393536567,
      "num_tokens": 2112517.0,
      "step": 1150
    },
    {
      "entropy": 5.968696737289429,
      "epoch": 0.0964751085867023,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999981317055116,
      "loss": 5.9811,
      "mean_token_accuracy": 0.13498256728053093,
      "num_tokens": 2121355.0,
      "step": 1155
    },
    {
      "entropy": 6.164517164230347,
      "epoch": 0.09689274974941531,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004999980084184244,
      "loss": 6.1112,
      "mean_token_accuracy": 0.1324291430413723,
      "num_tokens": 2129278.0,
      "step": 1160
    },
    {
      "entropy": 6.211386203765869,
      "epoch": 0.0973103909121283,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999978811924762,
      "loss": 6.2848,
      "mean_token_accuracy": 0.12834196388721467,
      "num_tokens": 2139548.0,
      "step": 1165
    },
    {
      "entropy": 6.049183940887451,
      "epoch": 0.09772803207484129,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999977500276693,
      "loss": 6.1727,
      "mean_token_accuracy": 0.12752494290471078,
      "num_tokens": 2148765.0,
      "step": 1170
    },
    {
      "entropy": 6.184184598922729,
      "epoch": 0.0981456732375543,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999976149240058,
      "loss": 6.0723,
      "mean_token_accuracy": 0.13661946654319762,
      "num_tokens": 2157482.0,
      "step": 1175
    },
    {
      "entropy": 6.077676248550415,
      "epoch": 0.09856331440026729,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999974758814884,
      "loss": 6.1873,
      "mean_token_accuracy": 0.12640749365091325,
      "num_tokens": 2167320.0,
      "step": 1180
    },
    {
      "entropy": 6.120995664596558,
      "epoch": 0.09898095556298028,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999973329001193,
      "loss": 6.1323,
      "mean_token_accuracy": 0.13020789697766305,
      "num_tokens": 2175454.0,
      "step": 1185
    },
    {
      "entropy": 6.151616096496582,
      "epoch": 0.09939859672569329,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999971859799009,
      "loss": 6.0682,
      "mean_token_accuracy": 0.13091856390237808,
      "num_tokens": 2184391.0,
      "step": 1190
    },
    {
      "entropy": 5.982958078384399,
      "epoch": 0.09981623788840628,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999970351208361,
      "loss": 5.9463,
      "mean_token_accuracy": 0.13676698952913285,
      "num_tokens": 2193309.0,
      "step": 1195
    },
    {
      "entropy": 6.130738019943237,
      "epoch": 0.10023387905111927,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999968803229275,
      "loss": 6.099,
      "mean_token_accuracy": 0.12918018624186517,
      "num_tokens": 2202057.0,
      "step": 1200
    },
    {
      "entropy": 6.048878574371338,
      "epoch": 0.10065152021383228,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999967215861775,
      "loss": 6.1954,
      "mean_token_accuracy": 0.12895920276641845,
      "num_tokens": 2212384.0,
      "step": 1205
    },
    {
      "entropy": 6.216175651550293,
      "epoch": 0.10106916137654527,
      "grad_norm": 1.09375,
      "learning_rate": 0.000499996558910589,
      "loss": 6.094,
      "mean_token_accuracy": 0.13314013928174973,
      "num_tokens": 2221451.0,
      "step": 1210
    },
    {
      "entropy": 5.950415802001953,
      "epoch": 0.10148680253925826,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000499996392296165,
      "loss": 6.0078,
      "mean_token_accuracy": 0.13325752466917037,
      "num_tokens": 2230440.0,
      "step": 1215
    },
    {
      "entropy": 6.147376155853271,
      "epoch": 0.10190444370197127,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999962217429083,
      "loss": 6.191,
      "mean_token_accuracy": 0.12644386813044547,
      "num_tokens": 2239985.0,
      "step": 1220
    },
    {
      "entropy": 6.079806327819824,
      "epoch": 0.10232208486468426,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004999960472508219,
      "loss": 6.05,
      "mean_token_accuracy": 0.1319073148071766,
      "num_tokens": 2250391.0,
      "step": 1225
    },
    {
      "entropy": 6.162016820907593,
      "epoch": 0.10273972602739725,
      "grad_norm": 1.015625,
      "learning_rate": 0.000499995868819909,
      "loss": 6.1621,
      "mean_token_accuracy": 0.12812372222542762,
      "num_tokens": 2260620.0,
      "step": 1230
    },
    {
      "entropy": 6.1506141185760494,
      "epoch": 0.10315736719011026,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004999956864501725,
      "loss": 6.1254,
      "mean_token_accuracy": 0.1344836674630642,
      "num_tokens": 2271176.0,
      "step": 1235
    },
    {
      "entropy": 6.071055316925049,
      "epoch": 0.10357500835282325,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999955001416156,
      "loss": 6.0197,
      "mean_token_accuracy": 0.1391293317079544,
      "num_tokens": 2280371.0,
      "step": 1240
    },
    {
      "entropy": 5.98853645324707,
      "epoch": 0.10399264951553625,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999953098942417,
      "loss": 5.9859,
      "mean_token_accuracy": 0.13849394246935845,
      "num_tokens": 2289015.0,
      "step": 1245
    },
    {
      "entropy": 6.034357261657715,
      "epoch": 0.10441029067824925,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999951157080541,
      "loss": 6.0661,
      "mean_token_accuracy": 0.13927728608250617,
      "num_tokens": 2297409.0,
      "step": 1250
    },
    {
      "entropy": 6.035771656036377,
      "epoch": 0.10482793184096224,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999949175830561,
      "loss": 6.0999,
      "mean_token_accuracy": 0.1318223387002945,
      "num_tokens": 2307017.0,
      "step": 1255
    },
    {
      "entropy": 6.02630467414856,
      "epoch": 0.10524557300367524,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999947155192514,
      "loss": 5.9257,
      "mean_token_accuracy": 0.14496939480304719,
      "num_tokens": 2316014.0,
      "step": 1260
    },
    {
      "entropy": 6.0644933700561525,
      "epoch": 0.10566321416638824,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999945095166433,
      "loss": 6.04,
      "mean_token_accuracy": 0.13692282512784004,
      "num_tokens": 2326326.0,
      "step": 1265
    },
    {
      "entropy": 6.156657695770264,
      "epoch": 0.10608085532910123,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999942995752354,
      "loss": 6.1305,
      "mean_token_accuracy": 0.13163789063692094,
      "num_tokens": 2335108.0,
      "step": 1270
    },
    {
      "entropy": 6.030630016326905,
      "epoch": 0.10649849649181423,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004999940856950315,
      "loss": 6.0516,
      "mean_token_accuracy": 0.14165214449167252,
      "num_tokens": 2343971.0,
      "step": 1275
    },
    {
      "entropy": 6.095973300933838,
      "epoch": 0.10691613765452723,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999938678760354,
      "loss": 6.1169,
      "mean_token_accuracy": 0.12894785180687904,
      "num_tokens": 2353742.0,
      "step": 1280
    },
    {
      "entropy": 6.044046449661255,
      "epoch": 0.10733377881724022,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999936461182507,
      "loss": 6.0673,
      "mean_token_accuracy": 0.14003410264849664,
      "num_tokens": 2363629.0,
      "step": 1285
    },
    {
      "entropy": 5.9364416122436525,
      "epoch": 0.10775141997995323,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999934204216814,
      "loss": 5.9119,
      "mean_token_accuracy": 0.14370162710547446,
      "num_tokens": 2372705.0,
      "step": 1290
    },
    {
      "entropy": 6.036751222610474,
      "epoch": 0.10816906114266622,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999931907863315,
      "loss": 6.0254,
      "mean_token_accuracy": 0.14308105707168578,
      "num_tokens": 2382136.0,
      "step": 1295
    },
    {
      "entropy": 5.983735466003418,
      "epoch": 0.10858670230537922,
      "grad_norm": 1.09375,
      "learning_rate": 0.000499992957212205,
      "loss": 6.0058,
      "mean_token_accuracy": 0.13456499874591826,
      "num_tokens": 2391060.0,
      "step": 1300
    },
    {
      "entropy": 6.100954914093018,
      "epoch": 0.10900434346809222,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999927196993059,
      "loss": 6.1417,
      "mean_token_accuracy": 0.13201688900589942,
      "num_tokens": 2399885.0,
      "step": 1305
    },
    {
      "entropy": 6.102160978317261,
      "epoch": 0.10942198463080521,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004999924782476384,
      "loss": 5.9898,
      "mean_token_accuracy": 0.1415901854634285,
      "num_tokens": 2409654.0,
      "step": 1310
    },
    {
      "entropy": 6.028567504882813,
      "epoch": 0.1098396257935182,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999922328572068,
      "loss": 6.0838,
      "mean_token_accuracy": 0.1362982250750065,
      "num_tokens": 2419342.0,
      "step": 1315
    },
    {
      "entropy": 5.955235624313355,
      "epoch": 0.11025726695623121,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999919835280154,
      "loss": 5.9448,
      "mean_token_accuracy": 0.141665992885828,
      "num_tokens": 2428845.0,
      "step": 1320
    },
    {
      "entropy": 6.083720779418945,
      "epoch": 0.1106749081189442,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999917302600684,
      "loss": 5.9526,
      "mean_token_accuracy": 0.14230969846248626,
      "num_tokens": 2437134.0,
      "step": 1325
    },
    {
      "entropy": 5.869806337356567,
      "epoch": 0.1110925492816572,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999914730533704,
      "loss": 5.7808,
      "mean_token_accuracy": 0.14912110045552254,
      "num_tokens": 2445438.0,
      "step": 1330
    },
    {
      "entropy": 5.9724204540252686,
      "epoch": 0.1115101904443702,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999912119079258,
      "loss": 6.1154,
      "mean_token_accuracy": 0.13888705000281334,
      "num_tokens": 2455039.0,
      "step": 1335
    },
    {
      "entropy": 6.091151475906372,
      "epoch": 0.1119278316070832,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004999909468237393,
      "loss": 6.0499,
      "mean_token_accuracy": 0.13516307920217513,
      "num_tokens": 2465170.0,
      "step": 1340
    },
    {
      "entropy": 5.9298077583312985,
      "epoch": 0.11234547276979619,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999906778008153,
      "loss": 5.8051,
      "mean_token_accuracy": 0.1507083296775818,
      "num_tokens": 2473688.0,
      "step": 1345
    },
    {
      "entropy": 5.883486986160278,
      "epoch": 0.11276311393250919,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999904048391588,
      "loss": 6.0757,
      "mean_token_accuracy": 0.1353744611144066,
      "num_tokens": 2482335.0,
      "step": 1350
    },
    {
      "entropy": 6.085854721069336,
      "epoch": 0.11318075509522219,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999901279387743,
      "loss": 5.9809,
      "mean_token_accuracy": 0.14239128828048705,
      "num_tokens": 2491808.0,
      "step": 1355
    },
    {
      "entropy": 6.159217739105225,
      "epoch": 0.11359839625793518,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999898470996669,
      "loss": 6.1325,
      "mean_token_accuracy": 0.1258760765194893,
      "num_tokens": 2501422.0,
      "step": 1360
    },
    {
      "entropy": 5.951807594299316,
      "epoch": 0.11401603742064818,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004999895623218415,
      "loss": 5.9229,
      "mean_token_accuracy": 0.14330672323703766,
      "num_tokens": 2511201.0,
      "step": 1365
    },
    {
      "entropy": 6.010347700119018,
      "epoch": 0.11443367858336118,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999892736053027,
      "loss": 5.9507,
      "mean_token_accuracy": 0.1380159690976143,
      "num_tokens": 2519798.0,
      "step": 1370
    },
    {
      "entropy": 5.919538259506226,
      "epoch": 0.11485131974607417,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999889809500561,
      "loss": 5.9262,
      "mean_token_accuracy": 0.1453268438577652,
      "num_tokens": 2529177.0,
      "step": 1375
    },
    {
      "entropy": 6.072020483016968,
      "epoch": 0.11526896090878717,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999886843561065,
      "loss": 5.9368,
      "mean_token_accuracy": 0.1449681594967842,
      "num_tokens": 2537571.0,
      "step": 1380
    },
    {
      "entropy": 5.872820711135864,
      "epoch": 0.11568660207150017,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999883838234592,
      "loss": 5.8471,
      "mean_token_accuracy": 0.14640183076262475,
      "num_tokens": 2546258.0,
      "step": 1385
    },
    {
      "entropy": 5.917314434051514,
      "epoch": 0.11610424323421316,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999880793521194,
      "loss": 5.871,
      "mean_token_accuracy": 0.13499426245689392,
      "num_tokens": 2555862.0,
      "step": 1390
    },
    {
      "entropy": 5.775463819503784,
      "epoch": 0.11652188439692616,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999877709420924,
      "loss": 5.8884,
      "mean_token_accuracy": 0.1478126809000969,
      "num_tokens": 2564583.0,
      "step": 1395
    },
    {
      "entropy": 5.995883417129517,
      "epoch": 0.11693952555963916,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999874585933837,
      "loss": 5.984,
      "mean_token_accuracy": 0.14109486490488052,
      "num_tokens": 2574217.0,
      "step": 1400
    },
    {
      "entropy": 5.964802837371826,
      "epoch": 0.11735716672235215,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999871423059987,
      "loss": 5.894,
      "mean_token_accuracy": 0.14622340500354766,
      "num_tokens": 2582821.0,
      "step": 1405
    },
    {
      "entropy": 5.935116100311279,
      "epoch": 0.11777480788506516,
      "grad_norm": 1.0,
      "learning_rate": 0.000499986822079943,
      "loss": 6.0564,
      "mean_token_accuracy": 0.1343017913401127,
      "num_tokens": 2592748.0,
      "step": 1410
    },
    {
      "entropy": 6.061121129989624,
      "epoch": 0.11819244904777815,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999864979152221,
      "loss": 5.9453,
      "mean_token_accuracy": 0.14890442043542862,
      "num_tokens": 2602053.0,
      "step": 1415
    },
    {
      "entropy": 5.839966821670532,
      "epoch": 0.11861009021049114,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999861698118417,
      "loss": 6.0227,
      "mean_token_accuracy": 0.13928924277424812,
      "num_tokens": 2610557.0,
      "step": 1420
    },
    {
      "entropy": 6.0450952529907225,
      "epoch": 0.11902773137320415,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999858377698076,
      "loss": 5.9138,
      "mean_token_accuracy": 0.14148174971342087,
      "num_tokens": 2619628.0,
      "step": 1425
    },
    {
      "entropy": 6.103807687759399,
      "epoch": 0.11944537253591714,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999855017891256,
      "loss": 6.0716,
      "mean_token_accuracy": 0.13505364283919336,
      "num_tokens": 2629284.0,
      "step": 1430
    },
    {
      "entropy": 5.925868225097656,
      "epoch": 0.11986301369863013,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999851618698017,
      "loss": 5.9931,
      "mean_token_accuracy": 0.14457735493779184,
      "num_tokens": 2638001.0,
      "step": 1435
    },
    {
      "entropy": 5.955863618850708,
      "epoch": 0.12028065486134314,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999848180118416,
      "loss": 5.9773,
      "mean_token_accuracy": 0.1491648070514202,
      "num_tokens": 2647071.0,
      "step": 1440
    },
    {
      "entropy": 6.025531244277954,
      "epoch": 0.12069829602405613,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999844702152516,
      "loss": 5.9622,
      "mean_token_accuracy": 0.1447582259774208,
      "num_tokens": 2655826.0,
      "step": 1445
    },
    {
      "entropy": 5.899601125717163,
      "epoch": 0.12111593718676913,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999841184800374,
      "loss": 5.8222,
      "mean_token_accuracy": 0.15455349087715148,
      "num_tokens": 2664431.0,
      "step": 1450
    },
    {
      "entropy": 5.779673719406128,
      "epoch": 0.12153357834948213,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999837628062055,
      "loss": 5.7455,
      "mean_token_accuracy": 0.15094979852437973,
      "num_tokens": 2673631.0,
      "step": 1455
    },
    {
      "entropy": 5.896817064285278,
      "epoch": 0.12195121951219512,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004999834031937621,
      "loss": 6.0289,
      "mean_token_accuracy": 0.14003881290555,
      "num_tokens": 2683363.0,
      "step": 1460
    },
    {
      "entropy": 6.015520000457764,
      "epoch": 0.12236886067490813,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004999830396427133,
      "loss": 5.979,
      "mean_token_accuracy": 0.1428418278694153,
      "num_tokens": 2692533.0,
      "step": 1465
    },
    {
      "entropy": 5.897101259231567,
      "epoch": 0.12278650183762112,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999826721530657,
      "loss": 5.9693,
      "mean_token_accuracy": 0.14741405099630356,
      "num_tokens": 2701211.0,
      "step": 1470
    },
    {
      "entropy": 6.052044248580932,
      "epoch": 0.12320414300033411,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999823007248255,
      "loss": 5.9683,
      "mean_token_accuracy": 0.14251434803009033,
      "num_tokens": 2709543.0,
      "step": 1475
    },
    {
      "entropy": 5.990947246551514,
      "epoch": 0.12362178416304712,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999819253579992,
      "loss": 5.973,
      "mean_token_accuracy": 0.14051907062530516,
      "num_tokens": 2718336.0,
      "step": 1480
    },
    {
      "entropy": 5.9257396221160885,
      "epoch": 0.12403942532576011,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999815460525938,
      "loss": 5.9216,
      "mean_token_accuracy": 0.14473359286785126,
      "num_tokens": 2726504.0,
      "step": 1485
    },
    {
      "entropy": 5.933391046524048,
      "epoch": 0.1244570664884731,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999811628086155,
      "loss": 5.9363,
      "mean_token_accuracy": 0.1432306505739689,
      "num_tokens": 2736076.0,
      "step": 1490
    },
    {
      "entropy": 5.884974098205566,
      "epoch": 0.1248747076511861,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000499980775626071,
      "loss": 5.8638,
      "mean_token_accuracy": 0.14747666344046592,
      "num_tokens": 2745497.0,
      "step": 1495
    },
    {
      "entropy": 5.883179140090943,
      "epoch": 0.12529234881389908,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999803845049674,
      "loss": 5.9851,
      "mean_token_accuracy": 0.14409718811511993,
      "num_tokens": 2754455.0,
      "step": 1500
    },
    {
      "entropy": 6.054201316833496,
      "epoch": 0.1257099899766121,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004999799894453112,
      "loss": 6.0131,
      "mean_token_accuracy": 0.14107876718044282,
      "num_tokens": 2763603.0,
      "step": 1505
    },
    {
      "entropy": 5.904917526245117,
      "epoch": 0.1261276311393251,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999795904471095,
      "loss": 5.8727,
      "mean_token_accuracy": 0.14535894691944123,
      "num_tokens": 2772152.0,
      "step": 1510
    },
    {
      "entropy": 6.038354587554932,
      "epoch": 0.12654527230203808,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004999791875103694,
      "loss": 5.9741,
      "mean_token_accuracy": 0.1499622419476509,
      "num_tokens": 2781459.0,
      "step": 1515
    },
    {
      "entropy": 5.9492429256439205,
      "epoch": 0.12696291346475108,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999787806350977,
      "loss": 5.9293,
      "mean_token_accuracy": 0.1426587760448456,
      "num_tokens": 2791177.0,
      "step": 1520
    },
    {
      "entropy": 5.861308526992798,
      "epoch": 0.1273805546274641,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999783698213016,
      "loss": 5.9155,
      "mean_token_accuracy": 0.15418764054775239,
      "num_tokens": 2800543.0,
      "step": 1525
    },
    {
      "entropy": 5.828676509857178,
      "epoch": 0.12779819579017707,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999779550689885,
      "loss": 5.7677,
      "mean_token_accuracy": 0.1490427389740944,
      "num_tokens": 2809301.0,
      "step": 1530
    },
    {
      "entropy": 5.975144577026367,
      "epoch": 0.12821583695289007,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999775363781654,
      "loss": 5.9798,
      "mean_token_accuracy": 0.1374480627477169,
      "num_tokens": 2818269.0,
      "step": 1535
    },
    {
      "entropy": 5.888175773620605,
      "epoch": 0.12863347811560308,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999771137488395,
      "loss": 5.8394,
      "mean_token_accuracy": 0.14905280917882918,
      "num_tokens": 2827170.0,
      "step": 1540
    },
    {
      "entropy": 5.831831455230713,
      "epoch": 0.12905111927831608,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999766871810188,
      "loss": 5.8317,
      "mean_token_accuracy": 0.15102441906929015,
      "num_tokens": 2835320.0,
      "step": 1545
    },
    {
      "entropy": 5.931498861312866,
      "epoch": 0.12946876044102906,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999762566747101,
      "loss": 5.9071,
      "mean_token_accuracy": 0.14569202587008476,
      "num_tokens": 2843755.0,
      "step": 1550
    },
    {
      "entropy": 6.002191972732544,
      "epoch": 0.12988640160374207,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999758222299214,
      "loss": 5.9512,
      "mean_token_accuracy": 0.13484849333763121,
      "num_tokens": 2853462.0,
      "step": 1555
    },
    {
      "entropy": 5.833185768127441,
      "epoch": 0.13030404276645507,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004999753838466601,
      "loss": 5.8504,
      "mean_token_accuracy": 0.14550600945949554,
      "num_tokens": 2863041.0,
      "step": 1560
    },
    {
      "entropy": 6.038375997543335,
      "epoch": 0.13072168392916805,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999749415249338,
      "loss": 6.0665,
      "mean_token_accuracy": 0.14647892266511917,
      "num_tokens": 2873019.0,
      "step": 1565
    },
    {
      "entropy": 5.836305475234985,
      "epoch": 0.13113932509188106,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999744952647505,
      "loss": 5.815,
      "mean_token_accuracy": 0.14850633218884468,
      "num_tokens": 2881467.0,
      "step": 1570
    },
    {
      "entropy": 5.894457483291626,
      "epoch": 0.13155696625459407,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999740450661176,
      "loss": 5.9128,
      "mean_token_accuracy": 0.14054032042622566,
      "num_tokens": 2890914.0,
      "step": 1575
    },
    {
      "entropy": 5.885600280761719,
      "epoch": 0.13197460741730704,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999735909290436,
      "loss": 5.8276,
      "mean_token_accuracy": 0.14965800568461418,
      "num_tokens": 2901164.0,
      "step": 1580
    },
    {
      "entropy": 5.899939250946045,
      "epoch": 0.13239224858002005,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999731328535358,
      "loss": 5.972,
      "mean_token_accuracy": 0.14218765646219253,
      "num_tokens": 2910560.0,
      "step": 1585
    },
    {
      "entropy": 5.862451267242432,
      "epoch": 0.13280988974273306,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999726708396026,
      "loss": 5.7887,
      "mean_token_accuracy": 0.15113321989774703,
      "num_tokens": 2918980.0,
      "step": 1590
    },
    {
      "entropy": 5.894169282913208,
      "epoch": 0.13322753090544603,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000499972204887252,
      "loss": 5.9799,
      "mean_token_accuracy": 0.14003576040267945,
      "num_tokens": 2929516.0,
      "step": 1595
    },
    {
      "entropy": 5.951022052764893,
      "epoch": 0.13364517206815904,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999717349964922,
      "loss": 6.0633,
      "mean_token_accuracy": 0.1349113292992115,
      "num_tokens": 2940603.0,
      "step": 1600
    },
    {
      "entropy": 5.994172143936157,
      "epoch": 0.13406281323087205,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999712611673313,
      "loss": 5.8962,
      "mean_token_accuracy": 0.14273757338523865,
      "num_tokens": 2950082.0,
      "step": 1605
    },
    {
      "entropy": 5.814873790740966,
      "epoch": 0.13448045439358502,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999707833997777,
      "loss": 5.7629,
      "mean_token_accuracy": 0.15739804357290268,
      "num_tokens": 2959098.0,
      "step": 1610
    },
    {
      "entropy": 5.9581045627594,
      "epoch": 0.13489809555629803,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999703016938398,
      "loss": 5.8869,
      "mean_token_accuracy": 0.14871252477169036,
      "num_tokens": 2968585.0,
      "step": 1615
    },
    {
      "entropy": 5.870134258270264,
      "epoch": 0.13531573671901104,
      "grad_norm": 0.9296875,
      "learning_rate": 0.000499969816049526,
      "loss": 5.9327,
      "mean_token_accuracy": 0.1441057324409485,
      "num_tokens": 2978430.0,
      "step": 1620
    },
    {
      "entropy": 6.060405111312866,
      "epoch": 0.13573337788172402,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004999693264668446,
      "loss": 5.9705,
      "mean_token_accuracy": 0.1415616549551487,
      "num_tokens": 2988398.0,
      "step": 1625
    },
    {
      "entropy": 5.9228941917419435,
      "epoch": 0.13615101904443702,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004999688329458045,
      "loss": 6.0189,
      "mean_token_accuracy": 0.1370128683745861,
      "num_tokens": 2999255.0,
      "step": 1630
    },
    {
      "entropy": 6.015878582000733,
      "epoch": 0.13656866020715003,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999683354864141,
      "loss": 5.8821,
      "mean_token_accuracy": 0.14219623506069184,
      "num_tokens": 3008638.0,
      "step": 1635
    },
    {
      "entropy": 5.79380989074707,
      "epoch": 0.136986301369863,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999678340886823,
      "loss": 5.893,
      "mean_token_accuracy": 0.14931328296661378,
      "num_tokens": 3018070.0,
      "step": 1640
    },
    {
      "entropy": 5.795630550384521,
      "epoch": 0.137403942532576,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004999673287526176,
      "loss": 5.8561,
      "mean_token_accuracy": 0.1456718251109123,
      "num_tokens": 3027878.0,
      "step": 1645
    },
    {
      "entropy": 6.055118751525879,
      "epoch": 0.13782158369528902,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999668194782291,
      "loss": 6.0385,
      "mean_token_accuracy": 0.14034320637583733,
      "num_tokens": 3037031.0,
      "step": 1650
    },
    {
      "entropy": 6.01564655303955,
      "epoch": 0.138239224858002,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004999663062655257,
      "loss": 5.9779,
      "mean_token_accuracy": 0.13667403906583786,
      "num_tokens": 3046311.0,
      "step": 1655
    },
    {
      "entropy": 5.877610874176026,
      "epoch": 0.138656866020715,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004999657891145162,
      "loss": 5.7878,
      "mean_token_accuracy": 0.14991068840026855,
      "num_tokens": 3055108.0,
      "step": 1660
    },
    {
      "entropy": 5.8218607902526855,
      "epoch": 0.139074507183428,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004999652680252099,
      "loss": 5.8356,
      "mean_token_accuracy": 0.15323519706726074,
      "num_tokens": 3063492.0,
      "step": 1665
    },
    {
      "entropy": 6.0089860439300535,
      "epoch": 0.139492148346141,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999647429976157,
      "loss": 5.8707,
      "mean_token_accuracy": 0.150077623128891,
      "num_tokens": 3072572.0,
      "step": 1670
    },
    {
      "entropy": 5.950778341293335,
      "epoch": 0.139909789508854,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004999642140317429,
      "loss": 5.9306,
      "mean_token_accuracy": 0.14366639852523805,
      "num_tokens": 3082942.0,
      "step": 1675
    },
    {
      "entropy": 5.84227180480957,
      "epoch": 0.140327430671567,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999636811276007,
      "loss": 5.8983,
      "mean_token_accuracy": 0.1497769057750702,
      "num_tokens": 3091410.0,
      "step": 1680
    },
    {
      "entropy": 6.025895690917968,
      "epoch": 0.14074507183427998,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999631442851986,
      "loss": 5.9669,
      "mean_token_accuracy": 0.1470497116446495,
      "num_tokens": 3100626.0,
      "step": 1685
    },
    {
      "entropy": 5.820451116561889,
      "epoch": 0.14116271299699298,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999626035045458,
      "loss": 5.763,
      "mean_token_accuracy": 0.15730037465691565,
      "num_tokens": 3109339.0,
      "step": 1690
    },
    {
      "entropy": 5.946425104141236,
      "epoch": 0.141580354159706,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999620587856519,
      "loss": 5.8886,
      "mean_token_accuracy": 0.14283772855997084,
      "num_tokens": 3118636.0,
      "step": 1695
    },
    {
      "entropy": 5.853169965744018,
      "epoch": 0.14199799532241897,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999615101285262,
      "loss": 5.8703,
      "mean_token_accuracy": 0.14250073656439782,
      "num_tokens": 3128162.0,
      "step": 1700
    },
    {
      "entropy": 5.882144975662231,
      "epoch": 0.14241563648513197,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999609575331786,
      "loss": 5.8669,
      "mean_token_accuracy": 0.14853606820106507,
      "num_tokens": 3136451.0,
      "step": 1705
    },
    {
      "entropy": 5.879884481430054,
      "epoch": 0.14283327764784498,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999604009996186,
      "loss": 5.8667,
      "mean_token_accuracy": 0.14338908046483995,
      "num_tokens": 3146599.0,
      "step": 1710
    },
    {
      "entropy": 5.972683095932007,
      "epoch": 0.14325091881055796,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000499959840527856,
      "loss": 5.9275,
      "mean_token_accuracy": 0.14261688515543938,
      "num_tokens": 3155941.0,
      "step": 1715
    },
    {
      "entropy": 5.865710830688476,
      "epoch": 0.14366855997327097,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999592761179007,
      "loss": 5.849,
      "mean_token_accuracy": 0.1503235414624214,
      "num_tokens": 3164641.0,
      "step": 1720
    },
    {
      "entropy": 5.8569518566131595,
      "epoch": 0.14408620113598397,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999587077697624,
      "loss": 5.8752,
      "mean_token_accuracy": 0.15404415428638457,
      "num_tokens": 3173701.0,
      "step": 1725
    },
    {
      "entropy": 5.823041582107544,
      "epoch": 0.14450384229869695,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999581354834512,
      "loss": 5.7338,
      "mean_token_accuracy": 0.15567242205142975,
      "num_tokens": 3182268.0,
      "step": 1730
    },
    {
      "entropy": 5.754219198226929,
      "epoch": 0.14492148346140996,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000499957559258977,
      "loss": 5.8158,
      "mean_token_accuracy": 0.14996908158063887,
      "num_tokens": 3191513.0,
      "step": 1735
    },
    {
      "entropy": 5.853840494155884,
      "epoch": 0.14533912462412296,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00049995697909635,
      "loss": 5.7625,
      "mean_token_accuracy": 0.14906580671668052,
      "num_tokens": 3200580.0,
      "step": 1740
    },
    {
      "entropy": 5.7556675434112545,
      "epoch": 0.14575676578683594,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999563949955803,
      "loss": 5.8603,
      "mean_token_accuracy": 0.15062788128852844,
      "num_tokens": 3209053.0,
      "step": 1745
    },
    {
      "entropy": 5.928824186325073,
      "epoch": 0.14617440694954895,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000499955806956678,
      "loss": 5.8441,
      "mean_token_accuracy": 0.15138443186879158,
      "num_tokens": 3218241.0,
      "step": 1750
    },
    {
      "entropy": 5.8904777526855465,
      "epoch": 0.14659204811226195,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999552149796536,
      "loss": 5.9059,
      "mean_token_accuracy": 0.14435181394219398,
      "num_tokens": 3227887.0,
      "step": 1755
    },
    {
      "entropy": 5.815938091278076,
      "epoch": 0.14700968927497493,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999546190645175,
      "loss": 5.7054,
      "mean_token_accuracy": 0.15759559869766235,
      "num_tokens": 3237281.0,
      "step": 1760
    },
    {
      "entropy": 5.811881685256958,
      "epoch": 0.14742733043768794,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999540192112799,
      "loss": 5.7542,
      "mean_token_accuracy": 0.14863720834255217,
      "num_tokens": 3246907.0,
      "step": 1765
    },
    {
      "entropy": 5.687792444229126,
      "epoch": 0.14784497160040094,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004999534154199514,
      "loss": 5.7613,
      "mean_token_accuracy": 0.14564004242420198,
      "num_tokens": 3256312.0,
      "step": 1770
    },
    {
      "entropy": 5.997764968872071,
      "epoch": 0.14826261276311392,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004999528076905426,
      "loss": 5.8538,
      "mean_token_accuracy": 0.14647961109876634,
      "num_tokens": 3265554.0,
      "step": 1775
    },
    {
      "entropy": 5.653895521163941,
      "epoch": 0.14868025392582693,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999521960230641,
      "loss": 5.701,
      "mean_token_accuracy": 0.15505964308977127,
      "num_tokens": 3274731.0,
      "step": 1780
    },
    {
      "entropy": 5.814744377136231,
      "epoch": 0.14909789508853993,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999515804175268,
      "loss": 5.8403,
      "mean_token_accuracy": 0.14269457161426544,
      "num_tokens": 3283518.0,
      "step": 1785
    },
    {
      "entropy": 5.923381757736206,
      "epoch": 0.1495155362512529,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000499950960873941,
      "loss": 5.8509,
      "mean_token_accuracy": 0.15064001828432083,
      "num_tokens": 3292244.0,
      "step": 1790
    },
    {
      "entropy": 5.874558019638061,
      "epoch": 0.14993317741396592,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999503373923182,
      "loss": 5.8117,
      "mean_token_accuracy": 0.15160613209009172,
      "num_tokens": 3301198.0,
      "step": 1795
    },
    {
      "entropy": 5.772550582885742,
      "epoch": 0.15035081857667892,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999497099726687,
      "loss": 5.8039,
      "mean_token_accuracy": 0.1560743421316147,
      "num_tokens": 3309983.0,
      "step": 1800
    },
    {
      "entropy": 5.89383111000061,
      "epoch": 0.1507684597393919,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004999490786150039,
      "loss": 5.88,
      "mean_token_accuracy": 0.15231931060552598,
      "num_tokens": 3319313.0,
      "step": 1805
    },
    {
      "entropy": 5.842616891860962,
      "epoch": 0.1511861009021049,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999484433193348,
      "loss": 5.7354,
      "mean_token_accuracy": 0.15028464645147324,
      "num_tokens": 3327497.0,
      "step": 1810
    },
    {
      "entropy": 5.748625373840332,
      "epoch": 0.15160374206481791,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999478040856722,
      "loss": 5.7535,
      "mean_token_accuracy": 0.15894282311201097,
      "num_tokens": 3337029.0,
      "step": 1815
    },
    {
      "entropy": 5.683879041671753,
      "epoch": 0.1520213832275309,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999471609140276,
      "loss": 5.7233,
      "mean_token_accuracy": 0.15257363468408586,
      "num_tokens": 3346243.0,
      "step": 1820
    },
    {
      "entropy": 5.895862579345703,
      "epoch": 0.1524390243902439,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999465138044122,
      "loss": 5.8765,
      "mean_token_accuracy": 0.14993367940187455,
      "num_tokens": 3354846.0,
      "step": 1825
    },
    {
      "entropy": 5.8310247421264645,
      "epoch": 0.1528566655529569,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999458627568373,
      "loss": 5.8289,
      "mean_token_accuracy": 0.14543151631951332,
      "num_tokens": 3364683.0,
      "step": 1830
    },
    {
      "entropy": 5.878261423110962,
      "epoch": 0.15327430671566988,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004999452077713144,
      "loss": 5.7708,
      "mean_token_accuracy": 0.15320614725351334,
      "num_tokens": 3374652.0,
      "step": 1835
    },
    {
      "entropy": 5.796768140792847,
      "epoch": 0.1536919478783829,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999445488478547,
      "loss": 5.7856,
      "mean_token_accuracy": 0.1481989949941635,
      "num_tokens": 3384040.0,
      "step": 1840
    },
    {
      "entropy": 5.8385844230651855,
      "epoch": 0.1541095890410959,
      "grad_norm": 1.171875,
      "learning_rate": 0.00049994388598647,
      "loss": 5.855,
      "mean_token_accuracy": 0.14915144294500352,
      "num_tokens": 3393185.0,
      "step": 1845
    },
    {
      "entropy": 5.814882135391235,
      "epoch": 0.15452723020380887,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999432191871719,
      "loss": 5.8005,
      "mean_token_accuracy": 0.15439388900995255,
      "num_tokens": 3403077.0,
      "step": 1850
    },
    {
      "entropy": 5.855295944213867,
      "epoch": 0.15494487136652188,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999425484499718,
      "loss": 5.7642,
      "mean_token_accuracy": 0.15828135311603547,
      "num_tokens": 3412165.0,
      "step": 1855
    },
    {
      "entropy": 5.789868879318237,
      "epoch": 0.1553625125292349,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999418737748817,
      "loss": 5.8162,
      "mean_token_accuracy": 0.1457912042737007,
      "num_tokens": 3421162.0,
      "step": 1860
    },
    {
      "entropy": 5.93405933380127,
      "epoch": 0.15578015369194786,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999411951619134,
      "loss": 5.9043,
      "mean_token_accuracy": 0.14147294089198112,
      "num_tokens": 3430363.0,
      "step": 1865
    },
    {
      "entropy": 5.850097751617431,
      "epoch": 0.15619779485466087,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999405126110786,
      "loss": 5.8087,
      "mean_token_accuracy": 0.15039258375763892,
      "num_tokens": 3439199.0,
      "step": 1870
    },
    {
      "entropy": 5.958965682983399,
      "epoch": 0.15661543601737388,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999398261223895,
      "loss": 5.9212,
      "mean_token_accuracy": 0.14575616419315338,
      "num_tokens": 3448720.0,
      "step": 1875
    },
    {
      "entropy": 5.840033674240113,
      "epoch": 0.15703307718008688,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999391356958579,
      "loss": 5.8919,
      "mean_token_accuracy": 0.14604691565036773,
      "num_tokens": 3458472.0,
      "step": 1880
    },
    {
      "entropy": 5.856070518493652,
      "epoch": 0.15745071834279986,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000499938441331496,
      "loss": 5.8284,
      "mean_token_accuracy": 0.14574070572853087,
      "num_tokens": 3467140.0,
      "step": 1885
    },
    {
      "entropy": 5.903375244140625,
      "epoch": 0.15786835950551287,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999377430293159,
      "loss": 5.817,
      "mean_token_accuracy": 0.1542054980993271,
      "num_tokens": 3476431.0,
      "step": 1890
    },
    {
      "entropy": 5.774544668197632,
      "epoch": 0.15828600066822587,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999370407893299,
      "loss": 5.7404,
      "mean_token_accuracy": 0.14861033111810684,
      "num_tokens": 3485768.0,
      "step": 1895
    },
    {
      "entropy": 5.885667562484741,
      "epoch": 0.15870364183093885,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999363346115501,
      "loss": 5.895,
      "mean_token_accuracy": 0.14465293884277344,
      "num_tokens": 3495678.0,
      "step": 1900
    },
    {
      "entropy": 5.832116174697876,
      "epoch": 0.15912128299365186,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004999356244959893,
      "loss": 5.8763,
      "mean_token_accuracy": 0.14869818314909936,
      "num_tokens": 3505168.0,
      "step": 1905
    },
    {
      "entropy": 5.854088401794433,
      "epoch": 0.15953892415636486,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004999349104426594,
      "loss": 5.83,
      "mean_token_accuracy": 0.14758925139904022,
      "num_tokens": 3514498.0,
      "step": 1910
    },
    {
      "entropy": 5.874787092208862,
      "epoch": 0.15995656531907784,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999341924515732,
      "loss": 5.7422,
      "mean_token_accuracy": 0.14831667989492417,
      "num_tokens": 3524617.0,
      "step": 1915
    },
    {
      "entropy": 5.707592535018921,
      "epoch": 0.16037420648179085,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999334705227431,
      "loss": 5.7259,
      "mean_token_accuracy": 0.1557302623987198,
      "num_tokens": 3533523.0,
      "step": 1920
    },
    {
      "entropy": 5.86808819770813,
      "epoch": 0.16079184764450385,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000499932744656182,
      "loss": 5.8106,
      "mean_token_accuracy": 0.148667673766613,
      "num_tokens": 3542922.0,
      "step": 1925
    },
    {
      "entropy": 5.842382860183716,
      "epoch": 0.16120948880721683,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999320148519024,
      "loss": 5.7929,
      "mean_token_accuracy": 0.15389788001775742,
      "num_tokens": 3551827.0,
      "step": 1930
    },
    {
      "entropy": 5.769865942001343,
      "epoch": 0.16162712996992984,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999312811099171,
      "loss": 5.764,
      "mean_token_accuracy": 0.15155478119850158,
      "num_tokens": 3560911.0,
      "step": 1935
    },
    {
      "entropy": 5.744738578796387,
      "epoch": 0.16204477113264285,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999305434302391,
      "loss": 5.7932,
      "mean_token_accuracy": 0.14804999828338622,
      "num_tokens": 3569473.0,
      "step": 1940
    },
    {
      "entropy": 5.809125232696533,
      "epoch": 0.16246241229535582,
      "grad_norm": 1.0625,
      "learning_rate": 0.000499929801812881,
      "loss": 5.7955,
      "mean_token_accuracy": 0.1538992702960968,
      "num_tokens": 3578843.0,
      "step": 1945
    },
    {
      "entropy": 5.8280191898345945,
      "epoch": 0.16288005345806883,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999290562578561,
      "loss": 5.7991,
      "mean_token_accuracy": 0.1495085433125496,
      "num_tokens": 3588304.0,
      "step": 1950
    },
    {
      "entropy": 5.762173223495483,
      "epoch": 0.16329769462078184,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999283067651773,
      "loss": 5.8568,
      "mean_token_accuracy": 0.14699322879314422,
      "num_tokens": 3598354.0,
      "step": 1955
    },
    {
      "entropy": 5.967223978042602,
      "epoch": 0.16371533578349481,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999275533348577,
      "loss": 5.8872,
      "mean_token_accuracy": 0.14445243775844574,
      "num_tokens": 3607503.0,
      "step": 1960
    },
    {
      "entropy": 5.690870523452759,
      "epoch": 0.16413297694620782,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999267959669106,
      "loss": 5.6962,
      "mean_token_accuracy": 0.15905150175094604,
      "num_tokens": 3617467.0,
      "step": 1965
    },
    {
      "entropy": 5.84067850112915,
      "epoch": 0.16455061810892083,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999260346613491,
      "loss": 5.7142,
      "mean_token_accuracy": 0.14913540482521057,
      "num_tokens": 3627048.0,
      "step": 1970
    },
    {
      "entropy": 5.721961498260498,
      "epoch": 0.1649682592716338,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999252694181867,
      "loss": 5.7598,
      "mean_token_accuracy": 0.1523315817117691,
      "num_tokens": 3636683.0,
      "step": 1975
    },
    {
      "entropy": 5.77244200706482,
      "epoch": 0.1653859004343468,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999245002374367,
      "loss": 5.6602,
      "mean_token_accuracy": 0.16195474863052367,
      "num_tokens": 3645231.0,
      "step": 1980
    },
    {
      "entropy": 5.743553972244262,
      "epoch": 0.16580354159705982,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999237271191125,
      "loss": 5.725,
      "mean_token_accuracy": 0.15497519820928574,
      "num_tokens": 3653851.0,
      "step": 1985
    },
    {
      "entropy": 5.739997816085816,
      "epoch": 0.1662211827597728,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999229500632279,
      "loss": 5.7126,
      "mean_token_accuracy": 0.153715518116951,
      "num_tokens": 3662720.0,
      "step": 1990
    },
    {
      "entropy": 5.651612234115601,
      "epoch": 0.1666388239224858,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004999221690697961,
      "loss": 5.6529,
      "mean_token_accuracy": 0.1546758532524109,
      "num_tokens": 3671777.0,
      "step": 1995
    },
    {
      "entropy": 5.773471879959106,
      "epoch": 0.1670564650851988,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999213841388312,
      "loss": 5.7392,
      "mean_token_accuracy": 0.1595829591155052,
      "num_tokens": 3681390.0,
      "step": 2000
    },
    {
      "entropy": 5.864161586761474,
      "epoch": 0.16747410624791179,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999205952703466,
      "loss": 5.7938,
      "mean_token_accuracy": 0.15263195633888244,
      "num_tokens": 3690899.0,
      "step": 2005
    },
    {
      "entropy": 5.887154245376587,
      "epoch": 0.1678917474106248,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999198024643563,
      "loss": 5.8879,
      "mean_token_accuracy": 0.13942136317491532,
      "num_tokens": 3699891.0,
      "step": 2010
    },
    {
      "entropy": 5.657262086868286,
      "epoch": 0.1683093885733378,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999190057208741,
      "loss": 5.6816,
      "mean_token_accuracy": 0.1632775440812111,
      "num_tokens": 3709860.0,
      "step": 2015
    },
    {
      "entropy": 5.774410057067871,
      "epoch": 0.16872702973605078,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000499918205039914,
      "loss": 5.8156,
      "mean_token_accuracy": 0.14972853809595107,
      "num_tokens": 3718896.0,
      "step": 2020
    },
    {
      "entropy": 5.787345123291016,
      "epoch": 0.16914467089876378,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999174004214899,
      "loss": 5.6125,
      "mean_token_accuracy": 0.15970027893781663,
      "num_tokens": 3727355.0,
      "step": 2025
    },
    {
      "entropy": 5.7231813907623295,
      "epoch": 0.1695623120614768,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999165918656161,
      "loss": 5.7293,
      "mean_token_accuracy": 0.14931901469826697,
      "num_tokens": 3736191.0,
      "step": 2030
    },
    {
      "entropy": 5.816195106506347,
      "epoch": 0.16997995322418977,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999157793723065,
      "loss": 5.7601,
      "mean_token_accuracy": 0.1600704610347748,
      "num_tokens": 3744832.0,
      "step": 2035
    },
    {
      "entropy": 5.696157789230346,
      "epoch": 0.17039759438690277,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999149629415755,
      "loss": 5.7095,
      "mean_token_accuracy": 0.14830667823553084,
      "num_tokens": 3753835.0,
      "step": 2040
    },
    {
      "entropy": 5.9111795902252195,
      "epoch": 0.17081523554961578,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999141425734374,
      "loss": 5.7839,
      "mean_token_accuracy": 0.1511392444372177,
      "num_tokens": 3762816.0,
      "step": 2045
    },
    {
      "entropy": 5.721945428848267,
      "epoch": 0.17123287671232876,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004999133182679064,
      "loss": 5.8427,
      "mean_token_accuracy": 0.1457610920071602,
      "num_tokens": 3773244.0,
      "step": 2050
    },
    {
      "entropy": 5.812287282943726,
      "epoch": 0.17165051787504176,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999124900249971,
      "loss": 5.6946,
      "mean_token_accuracy": 0.1568377897143364,
      "num_tokens": 3782122.0,
      "step": 2055
    },
    {
      "entropy": 5.745067119598389,
      "epoch": 0.17206815903775477,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999116578447239,
      "loss": 5.6901,
      "mean_token_accuracy": 0.16290647834539412,
      "num_tokens": 3791303.0,
      "step": 2060
    },
    {
      "entropy": 5.75031156539917,
      "epoch": 0.17248580020046775,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004999108217271015,
      "loss": 5.7535,
      "mean_token_accuracy": 0.1540791168808937,
      "num_tokens": 3800991.0,
      "step": 2065
    },
    {
      "entropy": 5.767945575714111,
      "epoch": 0.17290344136318075,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999099816721444,
      "loss": 5.6771,
      "mean_token_accuracy": 0.14992420375347137,
      "num_tokens": 3811293.0,
      "step": 2070
    },
    {
      "entropy": 5.840733480453491,
      "epoch": 0.17332108252589376,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999091376798673,
      "loss": 5.7827,
      "mean_token_accuracy": 0.15033572688698768,
      "num_tokens": 3820013.0,
      "step": 2075
    },
    {
      "entropy": 5.699107217788696,
      "epoch": 0.17373872368860674,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999082897502849,
      "loss": 5.6696,
      "mean_token_accuracy": 0.15458893924951553,
      "num_tokens": 3829099.0,
      "step": 2080
    },
    {
      "entropy": 5.677335739135742,
      "epoch": 0.17415636485131974,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999074378834123,
      "loss": 5.7674,
      "mean_token_accuracy": 0.15095923766493796,
      "num_tokens": 3838308.0,
      "step": 2085
    },
    {
      "entropy": 5.813047075271607,
      "epoch": 0.17457400601403275,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999065820792644,
      "loss": 5.7247,
      "mean_token_accuracy": 0.15711767971515656,
      "num_tokens": 3847650.0,
      "step": 2090
    },
    {
      "entropy": 5.866225051879883,
      "epoch": 0.17499164717674573,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999057223378559,
      "loss": 5.8998,
      "mean_token_accuracy": 0.14039052203297614,
      "num_tokens": 3857252.0,
      "step": 2095
    },
    {
      "entropy": 5.810293769836425,
      "epoch": 0.17540928833945874,
      "grad_norm": 1.078125,
      "learning_rate": 0.000499904858659202,
      "loss": 5.7376,
      "mean_token_accuracy": 0.1527712255716324,
      "num_tokens": 3865747.0,
      "step": 2100
    },
    {
      "entropy": 5.748090505599976,
      "epoch": 0.17582692950217174,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004999039910433179,
      "loss": 5.7179,
      "mean_token_accuracy": 0.1610593318939209,
      "num_tokens": 3874958.0,
      "step": 2105
    },
    {
      "entropy": 5.679863786697387,
      "epoch": 0.17624457066488472,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004999031194902187,
      "loss": 5.6593,
      "mean_token_accuracy": 0.16140231043100356,
      "num_tokens": 3885216.0,
      "step": 2110
    },
    {
      "entropy": 5.738522243499756,
      "epoch": 0.17666221182759773,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999022439999198,
      "loss": 5.7147,
      "mean_token_accuracy": 0.1608191043138504,
      "num_tokens": 3893973.0,
      "step": 2115
    },
    {
      "entropy": 5.77422251701355,
      "epoch": 0.17707985299031073,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999013645724362,
      "loss": 5.7932,
      "mean_token_accuracy": 0.1573753759264946,
      "num_tokens": 3903532.0,
      "step": 2120
    },
    {
      "entropy": 5.834164810180664,
      "epoch": 0.1774974941530237,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999004812077836,
      "loss": 5.708,
      "mean_token_accuracy": 0.15743714570999146,
      "num_tokens": 3913427.0,
      "step": 2125
    },
    {
      "entropy": 5.739658546447754,
      "epoch": 0.17791513531573672,
      "grad_norm": 1.0,
      "learning_rate": 0.0004998995939059773,
      "loss": 5.845,
      "mean_token_accuracy": 0.1429277092218399,
      "num_tokens": 3923870.0,
      "step": 2130
    },
    {
      "entropy": 5.769284439086914,
      "epoch": 0.17833277647844972,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000499898702667033,
      "loss": 5.6597,
      "mean_token_accuracy": 0.15647914707660676,
      "num_tokens": 3933469.0,
      "step": 2135
    },
    {
      "entropy": 5.6842976093292235,
      "epoch": 0.1787504176411627,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004998978074909661,
      "loss": 5.6779,
      "mean_token_accuracy": 0.1610175371170044,
      "num_tokens": 3942427.0,
      "step": 2140
    },
    {
      "entropy": 5.731369209289551,
      "epoch": 0.1791680588038757,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004998969083777925,
      "loss": 5.7213,
      "mean_token_accuracy": 0.1602757304906845,
      "num_tokens": 3951250.0,
      "step": 2145
    },
    {
      "entropy": 5.755114364624023,
      "epoch": 0.1795856999665887,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004998960053275277,
      "loss": 5.7129,
      "mean_token_accuracy": 0.15325510650873184,
      "num_tokens": 3959969.0,
      "step": 2150
    },
    {
      "entropy": 5.752820634841919,
      "epoch": 0.1800033411293017,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004998950983401875,
      "loss": 5.7425,
      "mean_token_accuracy": 0.15397500544786452,
      "num_tokens": 3968171.0,
      "step": 2155
    },
    {
      "entropy": 5.7606171607971195,
      "epoch": 0.1804209822920147,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004998941874157882,
      "loss": 5.6584,
      "mean_token_accuracy": 0.16023926734924315,
      "num_tokens": 3976686.0,
      "step": 2160
    },
    {
      "entropy": 5.6677477836608885,
      "epoch": 0.1808386234547277,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004998932725543453,
      "loss": 5.7763,
      "mean_token_accuracy": 0.14887330681085587,
      "num_tokens": 3986552.0,
      "step": 2165
    },
    {
      "entropy": 5.758628177642822,
      "epoch": 0.18125626461744068,
      "grad_norm": 1.125,
      "learning_rate": 0.0004998923537558749,
      "loss": 5.6363,
      "mean_token_accuracy": 0.15045822262763978,
      "num_tokens": 3995501.0,
      "step": 2170
    },
    {
      "entropy": 5.636182117462158,
      "epoch": 0.1816739057801537,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998914310203932,
      "loss": 5.62,
      "mean_token_accuracy": 0.15822093039751053,
      "num_tokens": 4004423.0,
      "step": 2175
    },
    {
      "entropy": 5.749627637863159,
      "epoch": 0.1820915469428667,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004998905043479163,
      "loss": 5.7937,
      "mean_token_accuracy": 0.15634801089763642,
      "num_tokens": 4013324.0,
      "step": 2180
    },
    {
      "entropy": 5.799656629562378,
      "epoch": 0.18250918810557967,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004998895737384605,
      "loss": 5.7072,
      "mean_token_accuracy": 0.1537037067115307,
      "num_tokens": 4021896.0,
      "step": 2185
    },
    {
      "entropy": 5.648739194869995,
      "epoch": 0.18292682926829268,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004998886391920418,
      "loss": 5.6855,
      "mean_token_accuracy": 0.15282756984233856,
      "num_tokens": 4030590.0,
      "step": 2190
    },
    {
      "entropy": 5.7324353694915775,
      "epoch": 0.18334447043100568,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004998877007086769,
      "loss": 5.5813,
      "mean_token_accuracy": 0.16229644566774368,
      "num_tokens": 4040041.0,
      "step": 2195
    },
    {
      "entropy": 5.735997533798217,
      "epoch": 0.18376211159371866,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004998867582883821,
      "loss": 5.7567,
      "mean_token_accuracy": 0.14448343887925147,
      "num_tokens": 4049119.0,
      "step": 2200
    },
    {
      "entropy": 5.758669328689575,
      "epoch": 0.18417975275643167,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004998858119311739,
      "loss": 5.6837,
      "mean_token_accuracy": 0.15333226174116135,
      "num_tokens": 4059265.0,
      "step": 2205
    },
    {
      "entropy": 5.7284187316894535,
      "epoch": 0.18459739391914468,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998848616370689,
      "loss": 5.7337,
      "mean_token_accuracy": 0.14927480071783067,
      "num_tokens": 4068363.0,
      "step": 2210
    },
    {
      "entropy": 5.653486967086792,
      "epoch": 0.18501503508185768,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004998839074060837,
      "loss": 5.6273,
      "mean_token_accuracy": 0.16272631734609605,
      "num_tokens": 4077390.0,
      "step": 2215
    },
    {
      "entropy": 5.7014764785766605,
      "epoch": 0.18543267624457066,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000499882949238235,
      "loss": 5.6943,
      "mean_token_accuracy": 0.14898738339543344,
      "num_tokens": 4087331.0,
      "step": 2220
    },
    {
      "entropy": 5.934384870529175,
      "epoch": 0.18585031740728367,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004998819871335396,
      "loss": 5.8722,
      "mean_token_accuracy": 0.14616972133517264,
      "num_tokens": 4096425.0,
      "step": 2225
    },
    {
      "entropy": 5.685419225692749,
      "epoch": 0.18626795856999667,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004998810210920142,
      "loss": 5.6401,
      "mean_token_accuracy": 0.15437570661306382,
      "num_tokens": 4106056.0,
      "step": 2230
    },
    {
      "entropy": 5.718805170059204,
      "epoch": 0.18668559973270965,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998800511136759,
      "loss": 5.7408,
      "mean_token_accuracy": 0.15399838015437126,
      "num_tokens": 4115133.0,
      "step": 2235
    },
    {
      "entropy": 5.710136127471924,
      "epoch": 0.18710324089542266,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004998790771985417,
      "loss": 5.6286,
      "mean_token_accuracy": 0.15800086557865142,
      "num_tokens": 4123630.0,
      "step": 2240
    },
    {
      "entropy": 5.682529258728027,
      "epoch": 0.18752088205813566,
      "grad_norm": 1.0,
      "learning_rate": 0.0004998780993466284,
      "loss": 5.7856,
      "mean_token_accuracy": 0.1551293760538101,
      "num_tokens": 4133337.0,
      "step": 2245
    },
    {
      "entropy": 5.789374923706054,
      "epoch": 0.18793852322084864,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004998771175579534,
      "loss": 5.7124,
      "mean_token_accuracy": 0.15884945541620255,
      "num_tokens": 4142240.0,
      "step": 2250
    },
    {
      "entropy": 5.731400394439698,
      "epoch": 0.18835616438356165,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004998761318325337,
      "loss": 5.6866,
      "mean_token_accuracy": 0.1549566388130188,
      "num_tokens": 4151297.0,
      "step": 2255
    },
    {
      "entropy": 5.672632312774658,
      "epoch": 0.18877380554627465,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998751421703868,
      "loss": 5.6457,
      "mean_token_accuracy": 0.15859557688236237,
      "num_tokens": 4160193.0,
      "step": 2260
    },
    {
      "entropy": 5.80642557144165,
      "epoch": 0.18919144670898763,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004998741485715297,
      "loss": 5.7881,
      "mean_token_accuracy": 0.1534851834177971,
      "num_tokens": 4169948.0,
      "step": 2265
    },
    {
      "entropy": 5.929843711853027,
      "epoch": 0.18960908787170064,
      "grad_norm": 1.0,
      "learning_rate": 0.00049987315103598,
      "loss": 5.8201,
      "mean_token_accuracy": 0.1431059531867504,
      "num_tokens": 4179981.0,
      "step": 2270
    },
    {
      "entropy": 5.721387052536011,
      "epoch": 0.19002672903441364,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004998721495637551,
      "loss": 5.8286,
      "mean_token_accuracy": 0.15135889053344725,
      "num_tokens": 4189172.0,
      "step": 2275
    },
    {
      "entropy": 5.771591997146606,
      "epoch": 0.19044437019712662,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004998711441548726,
      "loss": 5.6287,
      "mean_token_accuracy": 0.15496868789196014,
      "num_tokens": 4197312.0,
      "step": 2280
    },
    {
      "entropy": 5.775719499588012,
      "epoch": 0.19086201135983963,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004998701348093499,
      "loss": 5.7028,
      "mean_token_accuracy": 0.1541793242096901,
      "num_tokens": 4206828.0,
      "step": 2285
    },
    {
      "entropy": 5.774220943450928,
      "epoch": 0.19127965252255263,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000499869121527205,
      "loss": 5.6556,
      "mean_token_accuracy": 0.15856624841690065,
      "num_tokens": 4214835.0,
      "step": 2290
    },
    {
      "entropy": 5.697559452056884,
      "epoch": 0.1916972936852656,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004998681043084553,
      "loss": 5.8386,
      "mean_token_accuracy": 0.1496319055557251,
      "num_tokens": 4225119.0,
      "step": 2295
    },
    {
      "entropy": 5.692424154281616,
      "epoch": 0.19211493484797862,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004998670831531188,
      "loss": 5.6711,
      "mean_token_accuracy": 0.15529378354549409,
      "num_tokens": 4235257.0,
      "step": 2300
    },
    {
      "entropy": 5.735343933105469,
      "epoch": 0.19253257601069163,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004998660580612135,
      "loss": 5.754,
      "mean_token_accuracy": 0.1595138430595398,
      "num_tokens": 4244211.0,
      "step": 2305
    },
    {
      "entropy": 5.7877946376800535,
      "epoch": 0.1929502171734046,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004998650290327569,
      "loss": 5.7569,
      "mean_token_accuracy": 0.15316392183303834,
      "num_tokens": 4253149.0,
      "step": 2310
    },
    {
      "entropy": 5.677975702285766,
      "epoch": 0.1933678583361176,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998639960677675,
      "loss": 5.6596,
      "mean_token_accuracy": 0.15832507610321045,
      "num_tokens": 4262557.0,
      "step": 2315
    },
    {
      "entropy": 5.718808221817016,
      "epoch": 0.19378549949883062,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004998629591662631,
      "loss": 5.682,
      "mean_token_accuracy": 0.15558916032314302,
      "num_tokens": 4272881.0,
      "step": 2320
    },
    {
      "entropy": 5.675476217269898,
      "epoch": 0.1942031406615436,
      "grad_norm": 1.171875,
      "learning_rate": 0.000499861918328262,
      "loss": 5.6633,
      "mean_token_accuracy": 0.16360289007425308,
      "num_tokens": 4281158.0,
      "step": 2325
    },
    {
      "entropy": 5.677481555938721,
      "epoch": 0.1946207818242566,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004998608735537822,
      "loss": 5.6715,
      "mean_token_accuracy": 0.15835799872875214,
      "num_tokens": 4290319.0,
      "step": 2330
    },
    {
      "entropy": 5.709587240219117,
      "epoch": 0.1950384229869696,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004998598248428423,
      "loss": 5.7666,
      "mean_token_accuracy": 0.15781829208135606,
      "num_tokens": 4299276.0,
      "step": 2335
    },
    {
      "entropy": 5.818220853805542,
      "epoch": 0.19545606414968258,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998587721954604,
      "loss": 5.7341,
      "mean_token_accuracy": 0.1592143088579178,
      "num_tokens": 4308291.0,
      "step": 2340
    },
    {
      "entropy": 5.6828516006469725,
      "epoch": 0.1958737053123956,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004998577156116551,
      "loss": 5.6658,
      "mean_token_accuracy": 0.15610508769750595,
      "num_tokens": 4316654.0,
      "step": 2345
    },
    {
      "entropy": 5.653003692626953,
      "epoch": 0.1962913464751086,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004998566550914448,
      "loss": 5.6049,
      "mean_token_accuracy": 0.15679550021886826,
      "num_tokens": 4325911.0,
      "step": 2350
    },
    {
      "entropy": 5.785430240631103,
      "epoch": 0.19670898763782158,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004998555906348481,
      "loss": 5.794,
      "mean_token_accuracy": 0.1561383381485939,
      "num_tokens": 4334846.0,
      "step": 2355
    },
    {
      "entropy": 5.768541526794434,
      "epoch": 0.19712662880053458,
      "grad_norm": 1.125,
      "learning_rate": 0.0004998545222418836,
      "loss": 5.6889,
      "mean_token_accuracy": 0.1616716906428337,
      "num_tokens": 4343728.0,
      "step": 2360
    },
    {
      "entropy": 5.662745571136474,
      "epoch": 0.1975442699632476,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998534499125701,
      "loss": 5.6636,
      "mean_token_accuracy": 0.15771860778331756,
      "num_tokens": 4352774.0,
      "step": 2365
    },
    {
      "entropy": 5.819053602218628,
      "epoch": 0.19796191112596057,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004998523736469263,
      "loss": 5.7907,
      "mean_token_accuracy": 0.15671314746141435,
      "num_tokens": 4361663.0,
      "step": 2370
    },
    {
      "entropy": 5.743377780914306,
      "epoch": 0.19837955228867357,
      "grad_norm": 1.078125,
      "learning_rate": 0.000499851293444971,
      "loss": 5.6366,
      "mean_token_accuracy": 0.16553718596696854,
      "num_tokens": 4370278.0,
      "step": 2375
    },
    {
      "entropy": 5.745715761184693,
      "epoch": 0.19879719345138658,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004998502093067231,
      "loss": 5.6973,
      "mean_token_accuracy": 0.15904364138841628,
      "num_tokens": 4379010.0,
      "step": 2380
    },
    {
      "entropy": 5.693148374557495,
      "epoch": 0.19921483461409956,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004998491212322017,
      "loss": 5.6679,
      "mean_token_accuracy": 0.154359894990921,
      "num_tokens": 4388145.0,
      "step": 2385
    },
    {
      "entropy": 5.703994178771973,
      "epoch": 0.19963247577681256,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004998480292214258,
      "loss": 5.7985,
      "mean_token_accuracy": 0.15730815529823303,
      "num_tokens": 4396321.0,
      "step": 2390
    },
    {
      "entropy": 5.759179210662841,
      "epoch": 0.20005011693952557,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004998469332744144,
      "loss": 5.5769,
      "mean_token_accuracy": 0.16447094231843948,
      "num_tokens": 4404854.0,
      "step": 2395
    },
    {
      "entropy": 5.507405042648315,
      "epoch": 0.20046775810223855,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004998458333911869,
      "loss": 5.5746,
      "mean_token_accuracy": 0.16705760657787322,
      "num_tokens": 4413890.0,
      "step": 2400
    },
    {
      "entropy": 5.776613044738769,
      "epoch": 0.20088539926495155,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004998447295717624,
      "loss": 5.7294,
      "mean_token_accuracy": 0.15241889730095864,
      "num_tokens": 4423019.0,
      "step": 2405
    },
    {
      "entropy": 5.735891437530517,
      "epoch": 0.20130304042766456,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004998436218161602,
      "loss": 5.6779,
      "mean_token_accuracy": 0.15640403777360917,
      "num_tokens": 4432340.0,
      "step": 2410
    },
    {
      "entropy": 5.673480558395386,
      "epoch": 0.20172068159037754,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004998425101243999,
      "loss": 5.6509,
      "mean_token_accuracy": 0.15851726457476617,
      "num_tokens": 4441834.0,
      "step": 2415
    },
    {
      "entropy": 5.637769412994385,
      "epoch": 0.20213832275309054,
      "grad_norm": 1.0,
      "learning_rate": 0.0004998413944965007,
      "loss": 5.6713,
      "mean_token_accuracy": 0.16047401875257492,
      "num_tokens": 4450864.0,
      "step": 2420
    },
    {
      "entropy": 5.720356225967407,
      "epoch": 0.20255596391580355,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004998402749324823,
      "loss": 5.7449,
      "mean_token_accuracy": 0.15220133364200591,
      "num_tokens": 4460697.0,
      "step": 2425
    },
    {
      "entropy": 5.90509066581726,
      "epoch": 0.20297360507851653,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998391514323642,
      "loss": 5.7588,
      "mean_token_accuracy": 0.15590522438287735,
      "num_tokens": 4469926.0,
      "step": 2430
    },
    {
      "entropy": 5.611506032943725,
      "epoch": 0.20339124624122953,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004998380239961661,
      "loss": 5.6788,
      "mean_token_accuracy": 0.15944595634937286,
      "num_tokens": 4478818.0,
      "step": 2435
    },
    {
      "entropy": 5.822584009170532,
      "epoch": 0.20380888740394254,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004998368926239078,
      "loss": 5.6183,
      "mean_token_accuracy": 0.159269979596138,
      "num_tokens": 4487451.0,
      "step": 2440
    },
    {
      "entropy": 5.65903000831604,
      "epoch": 0.20422652856665552,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000499835757315609,
      "loss": 5.7008,
      "mean_token_accuracy": 0.1526777759194374,
      "num_tokens": 4497674.0,
      "step": 2445
    },
    {
      "entropy": 5.638183975219727,
      "epoch": 0.20464416972936852,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004998346180712898,
      "loss": 5.6441,
      "mean_token_accuracy": 0.1633292853832245,
      "num_tokens": 4507666.0,
      "step": 2450
    },
    {
      "entropy": 5.711228942871093,
      "epoch": 0.20506181089208153,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998334748909698,
      "loss": 5.7809,
      "mean_token_accuracy": 0.15110641419887544,
      "num_tokens": 4517409.0,
      "step": 2455
    },
    {
      "entropy": 5.68793215751648,
      "epoch": 0.2054794520547945,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004998323277746694,
      "loss": 5.5384,
      "mean_token_accuracy": 0.1622202515602112,
      "num_tokens": 4526495.0,
      "step": 2460
    },
    {
      "entropy": 5.698691320419312,
      "epoch": 0.20589709321750752,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004998311767224081,
      "loss": 5.6345,
      "mean_token_accuracy": 0.16896701455116273,
      "num_tokens": 4535240.0,
      "step": 2465
    },
    {
      "entropy": 5.68572301864624,
      "epoch": 0.20631473438022052,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004998300217342067,
      "loss": 5.7317,
      "mean_token_accuracy": 0.15531685054302216,
      "num_tokens": 4544716.0,
      "step": 2470
    },
    {
      "entropy": 5.664935541152954,
      "epoch": 0.2067323755429335,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004998288628100851,
      "loss": 5.6548,
      "mean_token_accuracy": 0.15997060686349868,
      "num_tokens": 4553867.0,
      "step": 2475
    },
    {
      "entropy": 5.734103202819824,
      "epoch": 0.2071500167056465,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004998276999500636,
      "loss": 5.7049,
      "mean_token_accuracy": 0.15193945467472075,
      "num_tokens": 4562529.0,
      "step": 2480
    },
    {
      "entropy": 5.626588201522827,
      "epoch": 0.2075676578683595,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004998265331541627,
      "loss": 5.4978,
      "mean_token_accuracy": 0.17455072104930877,
      "num_tokens": 4571041.0,
      "step": 2485
    },
    {
      "entropy": 5.593395090103149,
      "epoch": 0.2079852990310725,
      "grad_norm": 1.125,
      "learning_rate": 0.0004998253624224026,
      "loss": 5.6442,
      "mean_token_accuracy": 0.1611686021089554,
      "num_tokens": 4579914.0,
      "step": 2490
    },
    {
      "entropy": 5.68196291923523,
      "epoch": 0.2084029401937855,
      "grad_norm": 1.265625,
      "learning_rate": 0.000499824187754804,
      "loss": 5.7267,
      "mean_token_accuracy": 0.15876575261354448,
      "num_tokens": 4590277.0,
      "step": 2495
    },
    {
      "entropy": 5.762881946563721,
      "epoch": 0.2088205813564985,
      "grad_norm": 1.125,
      "learning_rate": 0.0004998230091513872,
      "loss": 5.6413,
      "mean_token_accuracy": 0.1610427275300026,
      "num_tokens": 4598719.0,
      "step": 2500
    },
    {
      "entropy": 5.598127794265747,
      "epoch": 0.20923822251921148,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004998218266121731,
      "loss": 5.6013,
      "mean_token_accuracy": 0.15835593938827514,
      "num_tokens": 4607804.0,
      "step": 2505
    },
    {
      "entropy": 5.5917142868042,
      "epoch": 0.2096558636819245,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004998206401371823,
      "loss": 5.5838,
      "mean_token_accuracy": 0.16652322113513945,
      "num_tokens": 4616529.0,
      "step": 2510
    },
    {
      "entropy": 5.690482568740845,
      "epoch": 0.2100735048446375,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004998194497264357,
      "loss": 5.6343,
      "mean_token_accuracy": 0.15835141241550446,
      "num_tokens": 4624545.0,
      "step": 2515
    },
    {
      "entropy": 5.662758302688599,
      "epoch": 0.21049114600735047,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004998182553799538,
      "loss": 5.6138,
      "mean_token_accuracy": 0.15359725803136826,
      "num_tokens": 4633316.0,
      "step": 2520
    },
    {
      "entropy": 5.617012453079224,
      "epoch": 0.21090878717006348,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004998170570977579,
      "loss": 5.5718,
      "mean_token_accuracy": 0.16380760073661804,
      "num_tokens": 4643353.0,
      "step": 2525
    },
    {
      "entropy": 5.613987588882447,
      "epoch": 0.21132642833277648,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004998158548798687,
      "loss": 5.6596,
      "mean_token_accuracy": 0.1520226702094078,
      "num_tokens": 4651998.0,
      "step": 2530
    },
    {
      "entropy": 5.712827730178833,
      "epoch": 0.21174406949548946,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004998146487263074,
      "loss": 5.653,
      "mean_token_accuracy": 0.16016076356172562,
      "num_tokens": 4659937.0,
      "step": 2535
    },
    {
      "entropy": 5.596396350860596,
      "epoch": 0.21216171065820247,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004998134386370951,
      "loss": 5.6389,
      "mean_token_accuracy": 0.15896535590291022,
      "num_tokens": 4669396.0,
      "step": 2540
    },
    {
      "entropy": 5.761886167526245,
      "epoch": 0.21257935182091547,
      "grad_norm": 1.0,
      "learning_rate": 0.0004998122246122529,
      "loss": 5.706,
      "mean_token_accuracy": 0.16040253043174743,
      "num_tokens": 4679320.0,
      "step": 2545
    },
    {
      "entropy": 5.712255525588989,
      "epoch": 0.21299699298362845,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004998110066518021,
      "loss": 5.6267,
      "mean_token_accuracy": 0.15740251243114473,
      "num_tokens": 4688807.0,
      "step": 2550
    },
    {
      "entropy": 5.680971097946167,
      "epoch": 0.21341463414634146,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000499809784755764,
      "loss": 5.6229,
      "mean_token_accuracy": 0.16529802978038788,
      "num_tokens": 4697819.0,
      "step": 2555
    },
    {
      "entropy": 5.612238550186158,
      "epoch": 0.21383227530905446,
      "grad_norm": 1.015625,
      "learning_rate": 0.00049980855892416,
      "loss": 5.6116,
      "mean_token_accuracy": 0.16319564878940582,
      "num_tokens": 4706931.0,
      "step": 2560
    },
    {
      "entropy": 5.657007837295533,
      "epoch": 0.21424991647176747,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004998073291570116,
      "loss": 5.5977,
      "mean_token_accuracy": 0.1614841639995575,
      "num_tokens": 4716262.0,
      "step": 2565
    },
    {
      "entropy": 5.657294845581054,
      "epoch": 0.21466755763448045,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998060954543404,
      "loss": 5.641,
      "mean_token_accuracy": 0.15932284742593766,
      "num_tokens": 4725415.0,
      "step": 2570
    },
    {
      "entropy": 5.598384952545166,
      "epoch": 0.21508519879719346,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004998048578161678,
      "loss": 5.6799,
      "mean_token_accuracy": 0.15921542495489122,
      "num_tokens": 4734580.0,
      "step": 2575
    },
    {
      "entropy": 5.70758843421936,
      "epoch": 0.21550283995990646,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004998036162425155,
      "loss": 5.5042,
      "mean_token_accuracy": 0.16997790932655335,
      "num_tokens": 4742966.0,
      "step": 2580
    },
    {
      "entropy": 5.624478483200074,
      "epoch": 0.21592048112261944,
      "grad_norm": 1.125,
      "learning_rate": 0.0004998023707334055,
      "loss": 5.6602,
      "mean_token_accuracy": 0.15674813762307166,
      "num_tokens": 4751696.0,
      "step": 2585
    },
    {
      "entropy": 5.702862167358399,
      "epoch": 0.21633812228533245,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998011212888592,
      "loss": 5.6996,
      "mean_token_accuracy": 0.16383157521486283,
      "num_tokens": 4760460.0,
      "step": 2590
    },
    {
      "entropy": 5.6970478057861325,
      "epoch": 0.21675576344804545,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004997998679088988,
      "loss": 5.6199,
      "mean_token_accuracy": 0.15811368972063064,
      "num_tokens": 4769004.0,
      "step": 2595
    },
    {
      "entropy": 5.615769100189209,
      "epoch": 0.21717340461075843,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004997986105935461,
      "loss": 5.6043,
      "mean_token_accuracy": 0.16909316480159758,
      "num_tokens": 4777204.0,
      "step": 2600
    },
    {
      "entropy": 5.713931846618652,
      "epoch": 0.21759104577347144,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004997973493428231,
      "loss": 5.7354,
      "mean_token_accuracy": 0.15746706277132033,
      "num_tokens": 4786766.0,
      "step": 2605
    },
    {
      "entropy": 5.630015087127686,
      "epoch": 0.21800868693618444,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004997960841567519,
      "loss": 5.6147,
      "mean_token_accuracy": 0.16642356812953948,
      "num_tokens": 4796676.0,
      "step": 2610
    },
    {
      "entropy": 5.836607217788696,
      "epoch": 0.21842632809889742,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004997948150353548,
      "loss": 5.6294,
      "mean_token_accuracy": 0.16387240290641786,
      "num_tokens": 4805514.0,
      "step": 2615
    },
    {
      "entropy": 5.636025714874267,
      "epoch": 0.21884396926161043,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004997935419786537,
      "loss": 5.7164,
      "mean_token_accuracy": 0.1554280623793602,
      "num_tokens": 4815487.0,
      "step": 2620
    },
    {
      "entropy": 5.725597667694092,
      "epoch": 0.21926161042432343,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004997922649866712,
      "loss": 5.6606,
      "mean_token_accuracy": 0.16378234028816224,
      "num_tokens": 4825044.0,
      "step": 2625
    },
    {
      "entropy": 5.588494157791137,
      "epoch": 0.2196792515870364,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004997909840594294,
      "loss": 5.4646,
      "mean_token_accuracy": 0.17271657586097716,
      "num_tokens": 4833875.0,
      "step": 2630
    },
    {
      "entropy": 5.5453134059906,
      "epoch": 0.22009689274974942,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004997896991969509,
      "loss": 5.6878,
      "mean_token_accuracy": 0.15461718291044235,
      "num_tokens": 4843317.0,
      "step": 2635
    },
    {
      "entropy": 5.717306423187256,
      "epoch": 0.22051453391246242,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004997884103992581,
      "loss": 5.6694,
      "mean_token_accuracy": 0.1513319656252861,
      "num_tokens": 4852514.0,
      "step": 2640
    },
    {
      "entropy": 5.713863706588745,
      "epoch": 0.2209321750751754,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004997871176663736,
      "loss": 5.621,
      "mean_token_accuracy": 0.16299737840890885,
      "num_tokens": 4861260.0,
      "step": 2645
    },
    {
      "entropy": 5.643141365051269,
      "epoch": 0.2213498162378884,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004997858209983201,
      "loss": 5.5676,
      "mean_token_accuracy": 0.15755053833127022,
      "num_tokens": 4869654.0,
      "step": 2650
    },
    {
      "entropy": 5.614552354812622,
      "epoch": 0.22176745740060141,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004997845203951202,
      "loss": 5.632,
      "mean_token_accuracy": 0.1524375632405281,
      "num_tokens": 4878311.0,
      "step": 2655
    },
    {
      "entropy": 5.764042854309082,
      "epoch": 0.2221850985633144,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004997832158567967,
      "loss": 5.6864,
      "mean_token_accuracy": 0.15463019385933877,
      "num_tokens": 4887251.0,
      "step": 2660
    },
    {
      "entropy": 5.66120228767395,
      "epoch": 0.2226027397260274,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004997819073833724,
      "loss": 5.5745,
      "mean_token_accuracy": 0.16053226739168167,
      "num_tokens": 4896528.0,
      "step": 2665
    },
    {
      "entropy": 5.586405324935913,
      "epoch": 0.2230203808887404,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004997805949748702,
      "loss": 5.5996,
      "mean_token_accuracy": 0.15948226004838945,
      "num_tokens": 4904962.0,
      "step": 2670
    },
    {
      "entropy": 5.630958700180054,
      "epoch": 0.22343802205145338,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004997792786313132,
      "loss": 5.6198,
      "mean_token_accuracy": 0.1612951412796974,
      "num_tokens": 4914735.0,
      "step": 2675
    },
    {
      "entropy": 5.482047605514526,
      "epoch": 0.2238556632141664,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004997779583527243,
      "loss": 5.4973,
      "mean_token_accuracy": 0.15745694637298585,
      "num_tokens": 4924271.0,
      "step": 2680
    },
    {
      "entropy": 5.763424968719482,
      "epoch": 0.2242733043768794,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004997766341391268,
      "loss": 5.7335,
      "mean_token_accuracy": 0.1580530509352684,
      "num_tokens": 4933928.0,
      "step": 2685
    },
    {
      "entropy": 5.563910961151123,
      "epoch": 0.22469094553959237,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004997753059905437,
      "loss": 5.4389,
      "mean_token_accuracy": 0.1698251634836197,
      "num_tokens": 4942714.0,
      "step": 2690
    },
    {
      "entropy": 5.477238178253174,
      "epoch": 0.22510858670230538,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004997739739069981,
      "loss": 5.574,
      "mean_token_accuracy": 0.16250405907630922,
      "num_tokens": 4952297.0,
      "step": 2695
    },
    {
      "entropy": 5.622715759277344,
      "epoch": 0.22552622786501839,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004997726378885138,
      "loss": 5.5178,
      "mean_token_accuracy": 0.16804625689983368,
      "num_tokens": 4960947.0,
      "step": 2700
    },
    {
      "entropy": 5.691080617904663,
      "epoch": 0.22594386902773136,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004997712979351137,
      "loss": 5.6715,
      "mean_token_accuracy": 0.1612107589840889,
      "num_tokens": 4969705.0,
      "step": 2705
    },
    {
      "entropy": 5.678754997253418,
      "epoch": 0.22636151019044437,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004997699540468216,
      "loss": 5.6377,
      "mean_token_accuracy": 0.1642414227128029,
      "num_tokens": 4979112.0,
      "step": 2710
    },
    {
      "entropy": 5.575213766098022,
      "epoch": 0.22677915135315738,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004997686062236609,
      "loss": 5.5573,
      "mean_token_accuracy": 0.16429246664047242,
      "num_tokens": 4987894.0,
      "step": 2715
    },
    {
      "entropy": 5.629120254516602,
      "epoch": 0.22719679251587035,
      "grad_norm": 1.0,
      "learning_rate": 0.0004997672544656553,
      "loss": 5.6969,
      "mean_token_accuracy": 0.15555485635995864,
      "num_tokens": 4997774.0,
      "step": 2720
    },
    {
      "entropy": 5.6939263343811035,
      "epoch": 0.22761443367858336,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004997658987728283,
      "loss": 5.6257,
      "mean_token_accuracy": 0.16721180230379104,
      "num_tokens": 5006183.0,
      "step": 2725
    },
    {
      "entropy": 5.66974105834961,
      "epoch": 0.22803207484129637,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004997645391452035,
      "loss": 5.6352,
      "mean_token_accuracy": 0.15832276195287703,
      "num_tokens": 5016909.0,
      "step": 2730
    },
    {
      "entropy": 5.732269906997681,
      "epoch": 0.22844971600400935,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000499763175582805,
      "loss": 5.6664,
      "mean_token_accuracy": 0.16458674222230912,
      "num_tokens": 5027326.0,
      "step": 2735
    },
    {
      "entropy": 5.662186813354492,
      "epoch": 0.22886735716672235,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004997618080856566,
      "loss": 5.6201,
      "mean_token_accuracy": 0.16315631568431854,
      "num_tokens": 5037043.0,
      "step": 2740
    },
    {
      "entropy": 5.532242393493652,
      "epoch": 0.22928499832943536,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004997604366537823,
      "loss": 5.6013,
      "mean_token_accuracy": 0.1606644257903099,
      "num_tokens": 5045992.0,
      "step": 2745
    },
    {
      "entropy": 5.615709590911865,
      "epoch": 0.22970263949214834,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004997590612872058,
      "loss": 5.5828,
      "mean_token_accuracy": 0.16492637246847153,
      "num_tokens": 5055568.0,
      "step": 2750
    },
    {
      "entropy": 5.723384141921997,
      "epoch": 0.23012028065486134,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004997576819859515,
      "loss": 5.6511,
      "mean_token_accuracy": 0.15917448103427886,
      "num_tokens": 5065512.0,
      "step": 2755
    },
    {
      "entropy": 5.514522075653076,
      "epoch": 0.23053792181757435,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004997562987500434,
      "loss": 5.5148,
      "mean_token_accuracy": 0.16921254098415375,
      "num_tokens": 5074776.0,
      "step": 2760
    },
    {
      "entropy": 5.603356647491455,
      "epoch": 0.23095556298028733,
      "grad_norm": 1.0,
      "learning_rate": 0.0004997549115795058,
      "loss": 5.5586,
      "mean_token_accuracy": 0.1597614198923111,
      "num_tokens": 5084743.0,
      "step": 2765
    },
    {
      "entropy": 5.788118314743042,
      "epoch": 0.23137320414300033,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000499753520474363,
      "loss": 5.7119,
      "mean_token_accuracy": 0.15703203529119492,
      "num_tokens": 5095387.0,
      "step": 2770
    },
    {
      "entropy": 5.602227401733399,
      "epoch": 0.23179084530571334,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004997521254346391,
      "loss": 5.6378,
      "mean_token_accuracy": 0.16656922772526742,
      "num_tokens": 5104628.0,
      "step": 2775
    },
    {
      "entropy": 5.685851812362671,
      "epoch": 0.23220848646842632,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004997507264603589,
      "loss": 5.5585,
      "mean_token_accuracy": 0.16496562957763672,
      "num_tokens": 5113169.0,
      "step": 2780
    },
    {
      "entropy": 5.598643827438354,
      "epoch": 0.23262612763113932,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004997493235515466,
      "loss": 5.6414,
      "mean_token_accuracy": 0.16375131905078888,
      "num_tokens": 5122294.0,
      "step": 2785
    },
    {
      "entropy": 5.584103536605835,
      "epoch": 0.23304376879385233,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004997479167082269,
      "loss": 5.4675,
      "mean_token_accuracy": 0.17088187783956527,
      "num_tokens": 5131191.0,
      "step": 2790
    },
    {
      "entropy": 5.608424329757691,
      "epoch": 0.2334614099565653,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004997465059304243,
      "loss": 5.6914,
      "mean_token_accuracy": 0.15928376764059066,
      "num_tokens": 5141974.0,
      "step": 2795
    },
    {
      "entropy": 5.69712815284729,
      "epoch": 0.23387905111927831,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004997450912181637,
      "loss": 5.6253,
      "mean_token_accuracy": 0.16184188574552535,
      "num_tokens": 5151412.0,
      "step": 2800
    },
    {
      "entropy": 5.600370121002197,
      "epoch": 0.23429669228199132,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004997436725714696,
      "loss": 5.5465,
      "mean_token_accuracy": 0.17141722291707992,
      "num_tokens": 5160989.0,
      "step": 2805
    },
    {
      "entropy": 5.604136896133423,
      "epoch": 0.2347143334447043,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004997422499903671,
      "loss": 5.6206,
      "mean_token_accuracy": 0.1575893685221672,
      "num_tokens": 5170241.0,
      "step": 2810
    },
    {
      "entropy": 5.762957906723022,
      "epoch": 0.2351319746074173,
      "grad_norm": 1.078125,
      "learning_rate": 0.000499740823474881,
      "loss": 5.7537,
      "mean_token_accuracy": 0.15024295151233674,
      "num_tokens": 5180247.0,
      "step": 2815
    },
    {
      "entropy": 5.622417020797729,
      "epoch": 0.2355496157701303,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004997393930250363,
      "loss": 5.631,
      "mean_token_accuracy": 0.16577300131320954,
      "num_tokens": 5189144.0,
      "step": 2820
    },
    {
      "entropy": 5.687289714813232,
      "epoch": 0.2359672569328433,
      "grad_norm": 1.03125,
      "learning_rate": 0.000499737958640858,
      "loss": 5.6849,
      "mean_token_accuracy": 0.15777652114629745,
      "num_tokens": 5199410.0,
      "step": 2825
    },
    {
      "entropy": 5.698432159423828,
      "epoch": 0.2363848980955563,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004997365203223711,
      "loss": 5.543,
      "mean_token_accuracy": 0.1669863536953926,
      "num_tokens": 5209032.0,
      "step": 2830
    },
    {
      "entropy": 5.460228633880615,
      "epoch": 0.2368025392582693,
      "grad_norm": 1.0625,
      "learning_rate": 0.000499735078069601,
      "loss": 5.4562,
      "mean_token_accuracy": 0.17202911972999574,
      "num_tokens": 5217787.0,
      "step": 2835
    },
    {
      "entropy": 5.605433750152588,
      "epoch": 0.23722018042098228,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004997336318825728,
      "loss": 5.5649,
      "mean_token_accuracy": 0.16585068553686141,
      "num_tokens": 5226962.0,
      "step": 2840
    },
    {
      "entropy": 5.561488389968872,
      "epoch": 0.23763782158369529,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004997321817613119,
      "loss": 5.54,
      "mean_token_accuracy": 0.16485002040863037,
      "num_tokens": 5236111.0,
      "step": 2845
    },
    {
      "entropy": 5.6743675708770756,
      "epoch": 0.2380554627464083,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004997307277058436,
      "loss": 5.6851,
      "mean_token_accuracy": 0.1568032205104828,
      "num_tokens": 5246628.0,
      "step": 2850
    },
    {
      "entropy": 5.619131994247437,
      "epoch": 0.23847310390912127,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004997292697161935,
      "loss": 5.6033,
      "mean_token_accuracy": 0.16125448048114777,
      "num_tokens": 5255543.0,
      "step": 2855
    },
    {
      "entropy": 5.53836874961853,
      "epoch": 0.23889074507183428,
      "grad_norm": 1.0625,
      "learning_rate": 0.000499727807792387,
      "loss": 5.4531,
      "mean_token_accuracy": 0.17420955300331115,
      "num_tokens": 5264559.0,
      "step": 2860
    },
    {
      "entropy": 5.477006959915161,
      "epoch": 0.23930838623454728,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004997263419344495,
      "loss": 5.5417,
      "mean_token_accuracy": 0.16187895089387894,
      "num_tokens": 5274599.0,
      "step": 2865
    },
    {
      "entropy": 5.608580589294434,
      "epoch": 0.23972602739726026,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004997248721424071,
      "loss": 5.5074,
      "mean_token_accuracy": 0.1679512843489647,
      "num_tokens": 5283495.0,
      "step": 2870
    },
    {
      "entropy": 5.635401773452759,
      "epoch": 0.24014366855997327,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004997233984162853,
      "loss": 5.5462,
      "mean_token_accuracy": 0.17000787556171418,
      "num_tokens": 5292064.0,
      "step": 2875
    },
    {
      "entropy": 5.577572917938232,
      "epoch": 0.24056130972268627,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004997219207561099,
      "loss": 5.5791,
      "mean_token_accuracy": 0.15797632038593293,
      "num_tokens": 5301696.0,
      "step": 2880
    },
    {
      "entropy": 5.710454320907592,
      "epoch": 0.24097895088539925,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004997204391619068,
      "loss": 5.6802,
      "mean_token_accuracy": 0.1591041773557663,
      "num_tokens": 5311427.0,
      "step": 2885
    },
    {
      "entropy": 5.741862344741821,
      "epoch": 0.24139659204811226,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004997189536337018,
      "loss": 5.6072,
      "mean_token_accuracy": 0.16055293530225753,
      "num_tokens": 5321406.0,
      "step": 2890
    },
    {
      "entropy": 5.633491802215576,
      "epoch": 0.24181423321082526,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004997174641715211,
      "loss": 5.6973,
      "mean_token_accuracy": 0.15916599482297897,
      "num_tokens": 5332257.0,
      "step": 2895
    },
    {
      "entropy": 5.550656461715699,
      "epoch": 0.24223187437353827,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004997159707753906,
      "loss": 5.4408,
      "mean_token_accuracy": 0.1674718514084816,
      "num_tokens": 5341857.0,
      "step": 2900
    },
    {
      "entropy": 5.60410509109497,
      "epoch": 0.24264951553625125,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004997144734453367,
      "loss": 5.5838,
      "mean_token_accuracy": 0.17222851514816284,
      "num_tokens": 5351367.0,
      "step": 2905
    },
    {
      "entropy": 5.458855724334716,
      "epoch": 0.24306715669896425,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004997129721813853,
      "loss": 5.4514,
      "mean_token_accuracy": 0.17036363631486892,
      "num_tokens": 5360627.0,
      "step": 2910
    },
    {
      "entropy": 5.609381294250488,
      "epoch": 0.24348479786167726,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004997114669835629,
      "loss": 5.5309,
      "mean_token_accuracy": 0.1677655428647995,
      "num_tokens": 5369703.0,
      "step": 2915
    },
    {
      "entropy": 5.63187084197998,
      "epoch": 0.24390243902439024,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004997099578518959,
      "loss": 5.5813,
      "mean_token_accuracy": 0.16637938767671584,
      "num_tokens": 5378432.0,
      "step": 2920
    },
    {
      "entropy": 5.485184192657471,
      "epoch": 0.24432008018710324,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004997084447864105,
      "loss": 5.4439,
      "mean_token_accuracy": 0.166954642534256,
      "num_tokens": 5386489.0,
      "step": 2925
    },
    {
      "entropy": 5.662143421173096,
      "epoch": 0.24473772134981625,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004997069277871333,
      "loss": 5.6627,
      "mean_token_accuracy": 0.1570824384689331,
      "num_tokens": 5395908.0,
      "step": 2930
    },
    {
      "entropy": 5.712649345397949,
      "epoch": 0.24515536251252923,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004997054068540908,
      "loss": 5.5795,
      "mean_token_accuracy": 0.16576846018433572,
      "num_tokens": 5404822.0,
      "step": 2935
    },
    {
      "entropy": 5.5368317604064945,
      "epoch": 0.24557300367524224,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004997038819873098,
      "loss": 5.4965,
      "mean_token_accuracy": 0.17085475027561187,
      "num_tokens": 5413980.0,
      "step": 2940
    },
    {
      "entropy": 5.604154777526856,
      "epoch": 0.24599064483795524,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004997023531868167,
      "loss": 5.5712,
      "mean_token_accuracy": 0.16370426267385482,
      "num_tokens": 5423538.0,
      "step": 2945
    },
    {
      "entropy": 5.5363991260528564,
      "epoch": 0.24640828600066822,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004997008204526386,
      "loss": 5.4912,
      "mean_token_accuracy": 0.16407241821289062,
      "num_tokens": 5431930.0,
      "step": 2950
    },
    {
      "entropy": 5.505523586273194,
      "epoch": 0.24682592716338123,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004996992837848021,
      "loss": 5.4793,
      "mean_token_accuracy": 0.17200600057840348,
      "num_tokens": 5440721.0,
      "step": 2955
    },
    {
      "entropy": 5.553699064254761,
      "epoch": 0.24724356832609423,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004996977431833342,
      "loss": 5.5793,
      "mean_token_accuracy": 0.16132830679416657,
      "num_tokens": 5449617.0,
      "step": 2960
    },
    {
      "entropy": 5.643174505233764,
      "epoch": 0.2476612094888072,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004996961986482618,
      "loss": 5.5937,
      "mean_token_accuracy": 0.1634842649102211,
      "num_tokens": 5458827.0,
      "step": 2965
    },
    {
      "entropy": 5.602717494964599,
      "epoch": 0.24807885065152022,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004996946501796119,
      "loss": 5.4708,
      "mean_token_accuracy": 0.16593398451805114,
      "num_tokens": 5467348.0,
      "step": 2970
    },
    {
      "entropy": 5.6201794147491455,
      "epoch": 0.24849649181423322,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004996930977774118,
      "loss": 5.5554,
      "mean_token_accuracy": 0.15850166752934455,
      "num_tokens": 5476645.0,
      "step": 2975
    },
    {
      "entropy": 5.603331804275513,
      "epoch": 0.2489141329769462,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004996915414416885,
      "loss": 5.6015,
      "mean_token_accuracy": 0.16163256764411926,
      "num_tokens": 5486120.0,
      "step": 2980
    },
    {
      "entropy": 5.723821020126342,
      "epoch": 0.2493317741396592,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004996899811724693,
      "loss": 5.5711,
      "mean_token_accuracy": 0.16415408551692962,
      "num_tokens": 5495313.0,
      "step": 2985
    },
    {
      "entropy": 5.556124019622803,
      "epoch": 0.2497494153023722,
      "grad_norm": 1.0,
      "learning_rate": 0.0004996884169697816,
      "loss": 5.5648,
      "mean_token_accuracy": 0.1665556788444519,
      "num_tokens": 5504587.0,
      "step": 2990
    },
    {
      "entropy": 5.50765814781189,
      "epoch": 0.2501670564650852,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004996868488336526,
      "loss": 5.5771,
      "mean_token_accuracy": 0.16290449798107148,
      "num_tokens": 5514188.0,
      "step": 2995
    },
    {
      "entropy": 5.618584823608399,
      "epoch": 0.25058469762779817,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004996852767641099,
      "loss": 5.6273,
      "mean_token_accuracy": 0.15422473549842836,
      "num_tokens": 5524075.0,
      "step": 3000
    },
    {
      "epoch": 0.25058469762779817,
      "eval_entropy": 5.509505016473965,
      "eval_loss": 5.556960582733154,
      "eval_mean_token_accuracy": 0.17263090768795453,
      "eval_num_tokens": 5524075.0,
      "eval_runtime": 27.5065,
      "eval_samples_per_second": 1372.876,
      "eval_steps_per_second": 171.632,
      "step": 3000
    },
    {
      "entropy": 5.569910383224487,
      "epoch": 0.2510023387905112,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004996837007611811,
      "loss": 5.4009,
      "mean_token_accuracy": 0.17270380705595018,
      "num_tokens": 5533323.0,
      "step": 3005
    },
    {
      "entropy": 5.512850713729859,
      "epoch": 0.2514199799532242,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004996821208248935,
      "loss": 5.4345,
      "mean_token_accuracy": 0.1709403544664383,
      "num_tokens": 5542288.0,
      "step": 3010
    },
    {
      "entropy": 5.643501949310303,
      "epoch": 0.25183762111593716,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004996805369552749,
      "loss": 5.5901,
      "mean_token_accuracy": 0.16013893485069275,
      "num_tokens": 5550869.0,
      "step": 3015
    },
    {
      "entropy": 5.553315210342407,
      "epoch": 0.2522552622786502,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004996789491523532,
      "loss": 5.518,
      "mean_token_accuracy": 0.17317127138376237,
      "num_tokens": 5560147.0,
      "step": 3020
    },
    {
      "entropy": 5.490715360641479,
      "epoch": 0.2526729034413632,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004996773574161559,
      "loss": 5.4083,
      "mean_token_accuracy": 0.17456791847944259,
      "num_tokens": 5568358.0,
      "step": 3025
    },
    {
      "entropy": 5.506763458251953,
      "epoch": 0.25309054460407615,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004996757617467112,
      "loss": 5.5345,
      "mean_token_accuracy": 0.1621170625090599,
      "num_tokens": 5577755.0,
      "step": 3030
    },
    {
      "entropy": 5.613587951660156,
      "epoch": 0.2535081857667892,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004996741621440466,
      "loss": 5.5781,
      "mean_token_accuracy": 0.15454234480857848,
      "num_tokens": 5587515.0,
      "step": 3035
    },
    {
      "entropy": 5.72170124053955,
      "epoch": 0.25392582692950216,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004996725586081906,
      "loss": 5.714,
      "mean_token_accuracy": 0.1539687216281891,
      "num_tokens": 5597860.0,
      "step": 3040
    },
    {
      "entropy": 5.637978744506836,
      "epoch": 0.25434346809221514,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004996709511391709,
      "loss": 5.5827,
      "mean_token_accuracy": 0.1627292513847351,
      "num_tokens": 5607302.0,
      "step": 3045
    },
    {
      "entropy": 5.61465163230896,
      "epoch": 0.2547611092549282,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004996693397370158,
      "loss": 5.6451,
      "mean_token_accuracy": 0.16155959963798522,
      "num_tokens": 5616345.0,
      "step": 3050
    },
    {
      "entropy": 5.569490194320679,
      "epoch": 0.25517875041764115,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004996677244017534,
      "loss": 5.4318,
      "mean_token_accuracy": 0.16576905995607377,
      "num_tokens": 5624745.0,
      "step": 3055
    },
    {
      "entropy": 5.563282489776611,
      "epoch": 0.25559639158035413,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004996661051334121,
      "loss": 5.5633,
      "mean_token_accuracy": 0.15848383605480193,
      "num_tokens": 5634022.0,
      "step": 3060
    },
    {
      "entropy": 5.600382471084595,
      "epoch": 0.25601403274306717,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004996644819320201,
      "loss": 5.5028,
      "mean_token_accuracy": 0.1660987839102745,
      "num_tokens": 5643631.0,
      "step": 3065
    },
    {
      "entropy": 5.613489151000977,
      "epoch": 0.25643167390578014,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004996628547976061,
      "loss": 5.5901,
      "mean_token_accuracy": 0.16259165555238725,
      "num_tokens": 5653155.0,
      "step": 3070
    },
    {
      "entropy": 5.604352855682373,
      "epoch": 0.2568493150684932,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004996612237301983,
      "loss": 5.6001,
      "mean_token_accuracy": 0.16388349682092668,
      "num_tokens": 5663091.0,
      "step": 3075
    },
    {
      "entropy": 5.540670013427734,
      "epoch": 0.25726695623120616,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004996595887298254,
      "loss": 5.493,
      "mean_token_accuracy": 0.17306645065546036,
      "num_tokens": 5671885.0,
      "step": 3080
    },
    {
      "entropy": 5.618255376815796,
      "epoch": 0.25768459739391913,
      "grad_norm": 1.15625,
      "learning_rate": 0.000499657949796516,
      "loss": 5.52,
      "mean_token_accuracy": 0.16467965692281722,
      "num_tokens": 5680225.0,
      "step": 3085
    },
    {
      "entropy": 5.434331464767456,
      "epoch": 0.25810223855663217,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004996563069302986,
      "loss": 5.3811,
      "mean_token_accuracy": 0.17379649877548217,
      "num_tokens": 5689456.0,
      "step": 3090
    },
    {
      "entropy": 5.517268896102905,
      "epoch": 0.25851987971934515,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004996546601312023,
      "loss": 5.4917,
      "mean_token_accuracy": 0.16643043011426925,
      "num_tokens": 5698127.0,
      "step": 3095
    },
    {
      "entropy": 5.621622753143311,
      "epoch": 0.2589375208820581,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004996530093992558,
      "loss": 5.545,
      "mean_token_accuracy": 0.17133604288101195,
      "num_tokens": 5706825.0,
      "step": 3100
    },
    {
      "entropy": 5.556325244903564,
      "epoch": 0.25935516204477116,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004996513547344877,
      "loss": 5.494,
      "mean_token_accuracy": 0.1657831236720085,
      "num_tokens": 5715677.0,
      "step": 3105
    },
    {
      "entropy": 5.503501224517822,
      "epoch": 0.25977280320748414,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004996496961369274,
      "loss": 5.4165,
      "mean_token_accuracy": 0.17088277041912078,
      "num_tokens": 5725371.0,
      "step": 3110
    },
    {
      "entropy": 5.553598070144654,
      "epoch": 0.2601904443701971,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004996480336066037,
      "loss": 5.521,
      "mean_token_accuracy": 0.1708841696381569,
      "num_tokens": 5734664.0,
      "step": 3115
    },
    {
      "entropy": 5.573401355743409,
      "epoch": 0.26060808553291015,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004996463671435458,
      "loss": 5.5226,
      "mean_token_accuracy": 0.1694071963429451,
      "num_tokens": 5743870.0,
      "step": 3120
    },
    {
      "entropy": 5.507350969314575,
      "epoch": 0.26102572669562313,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004996446967477829,
      "loss": 5.5359,
      "mean_token_accuracy": 0.16690430790185928,
      "num_tokens": 5752855.0,
      "step": 3125
    },
    {
      "entropy": 5.639628314971924,
      "epoch": 0.2614433678583361,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004996430224193441,
      "loss": 5.5937,
      "mean_token_accuracy": 0.1663447767496109,
      "num_tokens": 5762435.0,
      "step": 3130
    },
    {
      "entropy": 5.607776641845703,
      "epoch": 0.26186100902104914,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004996413441582589,
      "loss": 5.5813,
      "mean_token_accuracy": 0.16195459812879562,
      "num_tokens": 5771463.0,
      "step": 3135
    },
    {
      "entropy": 5.624371004104614,
      "epoch": 0.2622786501837621,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004996396619645564,
      "loss": 5.5852,
      "mean_token_accuracy": 0.15886165350675582,
      "num_tokens": 5780530.0,
      "step": 3140
    },
    {
      "entropy": 5.553997278213501,
      "epoch": 0.2626962913464751,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004996379758382662,
      "loss": 5.4534,
      "mean_token_accuracy": 0.17759004533290862,
      "num_tokens": 5789199.0,
      "step": 3145
    },
    {
      "entropy": 5.69573564529419,
      "epoch": 0.26311393250918813,
      "grad_norm": 0.96875,
      "learning_rate": 0.000499636285779418,
      "loss": 5.65,
      "mean_token_accuracy": 0.16340592503547668,
      "num_tokens": 5799170.0,
      "step": 3150
    },
    {
      "entropy": 5.620092391967773,
      "epoch": 0.2635315736719011,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004996345917880412,
      "loss": 5.5493,
      "mean_token_accuracy": 0.16142655313014984,
      "num_tokens": 5807662.0,
      "step": 3155
    },
    {
      "entropy": 5.4721107006073,
      "epoch": 0.2639492148346141,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004996328938641654,
      "loss": 5.4585,
      "mean_token_accuracy": 0.16569848209619523,
      "num_tokens": 5815790.0,
      "step": 3160
    },
    {
      "entropy": 5.6283488273620605,
      "epoch": 0.2643668559973271,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004996311920078204,
      "loss": 5.4755,
      "mean_token_accuracy": 0.17084998786449432,
      "num_tokens": 5824746.0,
      "step": 3165
    },
    {
      "entropy": 5.486237382888794,
      "epoch": 0.2647844971600401,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004996294862190361,
      "loss": 5.4854,
      "mean_token_accuracy": 0.1715046674013138,
      "num_tokens": 5833678.0,
      "step": 3170
    },
    {
      "entropy": 5.528229284286499,
      "epoch": 0.2652021383227531,
      "grad_norm": 1.15625,
      "learning_rate": 0.000499627776497842,
      "loss": 5.5334,
      "mean_token_accuracy": 0.16838407665491104,
      "num_tokens": 5842218.0,
      "step": 3175
    },
    {
      "entropy": 5.754685544967652,
      "epoch": 0.2656197794854661,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004996260628442686,
      "loss": 5.713,
      "mean_token_accuracy": 0.15619731843471527,
      "num_tokens": 5853250.0,
      "step": 3180
    },
    {
      "entropy": 5.566620397567749,
      "epoch": 0.2660374206481791,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004996243452583454,
      "loss": 5.4545,
      "mean_token_accuracy": 0.17067957371473313,
      "num_tokens": 5862661.0,
      "step": 3185
    },
    {
      "entropy": 5.601086616516113,
      "epoch": 0.26645506181089207,
      "grad_norm": 1.0,
      "learning_rate": 0.0004996226237401028,
      "loss": 5.6168,
      "mean_token_accuracy": 0.1654422849416733,
      "num_tokens": 5872327.0,
      "step": 3190
    },
    {
      "entropy": 5.52212347984314,
      "epoch": 0.2668727029736051,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004996208982895706,
      "loss": 5.4503,
      "mean_token_accuracy": 0.16680678874254226,
      "num_tokens": 5881539.0,
      "step": 3195
    },
    {
      "entropy": 5.622729396820068,
      "epoch": 0.2672903441363181,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004996191689067794,
      "loss": 5.5614,
      "mean_token_accuracy": 0.1628652647137642,
      "num_tokens": 5891602.0,
      "step": 3200
    },
    {
      "entropy": 5.5512511253356935,
      "epoch": 0.26770798529903106,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004996174355917591,
      "loss": 5.5499,
      "mean_token_accuracy": 0.1703685149550438,
      "num_tokens": 5899831.0,
      "step": 3205
    },
    {
      "entropy": 5.700180721282959,
      "epoch": 0.2681256264617441,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004996156983445404,
      "loss": 5.666,
      "mean_token_accuracy": 0.15713129043579102,
      "num_tokens": 5909193.0,
      "step": 3210
    },
    {
      "entropy": 5.683463907241821,
      "epoch": 0.26854326762445707,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004996139571651534,
      "loss": 5.5554,
      "mean_token_accuracy": 0.1625669240951538,
      "num_tokens": 5919353.0,
      "step": 3215
    },
    {
      "entropy": 5.471538925170899,
      "epoch": 0.26896090878717005,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004996122120536288,
      "loss": 5.5043,
      "mean_token_accuracy": 0.1690175250172615,
      "num_tokens": 5927853.0,
      "step": 3220
    },
    {
      "entropy": 5.503655052185058,
      "epoch": 0.2693785499498831,
      "grad_norm": 1.0,
      "learning_rate": 0.0004996104630099969,
      "loss": 5.6065,
      "mean_token_accuracy": 0.1630278766155243,
      "num_tokens": 5937943.0,
      "step": 3225
    },
    {
      "entropy": 5.63100643157959,
      "epoch": 0.26979619111259606,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004996087100342885,
      "loss": 5.4864,
      "mean_token_accuracy": 0.17056625485420226,
      "num_tokens": 5946213.0,
      "step": 3230
    },
    {
      "entropy": 5.515283298492432,
      "epoch": 0.27021383227530904,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004996069531265345,
      "loss": 5.4122,
      "mean_token_accuracy": 0.1662533089518547,
      "num_tokens": 5954997.0,
      "step": 3235
    },
    {
      "entropy": 5.515188264846802,
      "epoch": 0.2706314734380221,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004996051922867653,
      "loss": 5.4425,
      "mean_token_accuracy": 0.17493584454059602,
      "num_tokens": 5963947.0,
      "step": 3240
    },
    {
      "entropy": 5.549450969696045,
      "epoch": 0.27104911460073505,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004996034275150118,
      "loss": 5.4948,
      "mean_token_accuracy": 0.1700539380311966,
      "num_tokens": 5972077.0,
      "step": 3245
    },
    {
      "entropy": 5.467683124542236,
      "epoch": 0.27146675576344803,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000499601658811305,
      "loss": 5.4905,
      "mean_token_accuracy": 0.1693744868040085,
      "num_tokens": 5982649.0,
      "step": 3250
    },
    {
      "entropy": 5.608425951004028,
      "epoch": 0.27188439692616106,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004995998861756758,
      "loss": 5.5166,
      "mean_token_accuracy": 0.16742232888936998,
      "num_tokens": 5991349.0,
      "step": 3255
    },
    {
      "entropy": 5.506754159927368,
      "epoch": 0.27230203808887404,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004995981096081552,
      "loss": 5.4893,
      "mean_token_accuracy": 0.17103740125894545,
      "num_tokens": 6000555.0,
      "step": 3260
    },
    {
      "entropy": 5.731655216217041,
      "epoch": 0.272719679251587,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004995963291087744,
      "loss": 5.6098,
      "mean_token_accuracy": 0.1574689731001854,
      "num_tokens": 6009554.0,
      "step": 3265
    },
    {
      "entropy": 5.605317401885986,
      "epoch": 0.27313732041430006,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004995945446775645,
      "loss": 5.7056,
      "mean_token_accuracy": 0.15773640125989913,
      "num_tokens": 6019330.0,
      "step": 3270
    },
    {
      "entropy": 5.691388893127441,
      "epoch": 0.27355496157701303,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004995927563145566,
      "loss": 5.6377,
      "mean_token_accuracy": 0.16591110527515412,
      "num_tokens": 6028527.0,
      "step": 3275
    },
    {
      "entropy": 5.584264707565308,
      "epoch": 0.273972602739726,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004995909640197825,
      "loss": 5.5009,
      "mean_token_accuracy": 0.17068353593349456,
      "num_tokens": 6038223.0,
      "step": 3280
    },
    {
      "entropy": 5.459718418121338,
      "epoch": 0.27439024390243905,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004995891677932729,
      "loss": 5.4431,
      "mean_token_accuracy": 0.1723234698176384,
      "num_tokens": 6047459.0,
      "step": 3285
    },
    {
      "entropy": 5.680184316635132,
      "epoch": 0.274807885065152,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004995873676350597,
      "loss": 5.6245,
      "mean_token_accuracy": 0.16481543332338333,
      "num_tokens": 6057322.0,
      "step": 3290
    },
    {
      "entropy": 5.61651577949524,
      "epoch": 0.275225526227865,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004995855635451743,
      "loss": 5.7453,
      "mean_token_accuracy": 0.15445435047149658,
      "num_tokens": 6067520.0,
      "step": 3295
    },
    {
      "entropy": 5.5944427967071535,
      "epoch": 0.27564316739057804,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004995837555236484,
      "loss": 5.6221,
      "mean_token_accuracy": 0.16172602623701096,
      "num_tokens": 6076729.0,
      "step": 3300
    },
    {
      "entropy": 5.6910748958587645,
      "epoch": 0.276060808553291,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004995819435705133,
      "loss": 5.5689,
      "mean_token_accuracy": 0.1596569076180458,
      "num_tokens": 6086529.0,
      "step": 3305
    },
    {
      "entropy": 5.6409749507904055,
      "epoch": 0.276478449716004,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004995801276858011,
      "loss": 5.5777,
      "mean_token_accuracy": 0.1702018052339554,
      "num_tokens": 6095784.0,
      "step": 3310
    },
    {
      "entropy": 5.68081955909729,
      "epoch": 0.276896090878717,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004995783078695434,
      "loss": 5.5926,
      "mean_token_accuracy": 0.16117540150880813,
      "num_tokens": 6104626.0,
      "step": 3315
    },
    {
      "entropy": 5.550822114944458,
      "epoch": 0.27731373204143,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000499576484121772,
      "loss": 5.5063,
      "mean_token_accuracy": 0.17098721265792846,
      "num_tokens": 6113758.0,
      "step": 3320
    },
    {
      "entropy": 5.594186162948608,
      "epoch": 0.277731373204143,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004995746564425191,
      "loss": 5.5469,
      "mean_token_accuracy": 0.16868577748537064,
      "num_tokens": 6123265.0,
      "step": 3325
    },
    {
      "entropy": 5.6335364818573,
      "epoch": 0.278149014366856,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004995728248318163,
      "loss": 5.5909,
      "mean_token_accuracy": 0.1608134001493454,
      "num_tokens": 6132855.0,
      "step": 3330
    },
    {
      "entropy": 5.496687221527099,
      "epoch": 0.278566655529569,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004995709892896961,
      "loss": 5.4769,
      "mean_token_accuracy": 0.1676080346107483,
      "num_tokens": 6142739.0,
      "step": 3335
    },
    {
      "entropy": 5.513911485671997,
      "epoch": 0.278984296692282,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004995691498161903,
      "loss": 5.4272,
      "mean_token_accuracy": 0.17005877643823625,
      "num_tokens": 6152169.0,
      "step": 3340
    },
    {
      "entropy": 5.4408985614776615,
      "epoch": 0.279401937854995,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004995673064113313,
      "loss": 5.4454,
      "mean_token_accuracy": 0.1737777844071388,
      "num_tokens": 6160235.0,
      "step": 3345
    },
    {
      "entropy": 5.525840425491333,
      "epoch": 0.279819579017708,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004995654590751513,
      "loss": 5.4355,
      "mean_token_accuracy": 0.1697912961244583,
      "num_tokens": 6169440.0,
      "step": 3350
    },
    {
      "entropy": 5.500869989395142,
      "epoch": 0.28023722018042097,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004995636078076825,
      "loss": 5.4715,
      "mean_token_accuracy": 0.17083757370710373,
      "num_tokens": 6178456.0,
      "step": 3355
    },
    {
      "entropy": 5.51913628578186,
      "epoch": 0.280654861343134,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004995617526089575,
      "loss": 5.5321,
      "mean_token_accuracy": 0.16220356673002242,
      "num_tokens": 6186844.0,
      "step": 3360
    },
    {
      "entropy": 5.555132722854614,
      "epoch": 0.281072502505847,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004995598934790087,
      "loss": 5.4516,
      "mean_token_accuracy": 0.17172517627477646,
      "num_tokens": 6196117.0,
      "step": 3365
    },
    {
      "entropy": 5.563861608505249,
      "epoch": 0.28149014366855996,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004995580304178687,
      "loss": 5.4401,
      "mean_token_accuracy": 0.17228131592273713,
      "num_tokens": 6204636.0,
      "step": 3370
    },
    {
      "entropy": 5.4404088973999025,
      "epoch": 0.281907784831273,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004995561634255701,
      "loss": 5.4745,
      "mean_token_accuracy": 0.16716700941324233,
      "num_tokens": 6214537.0,
      "step": 3375
    },
    {
      "entropy": 5.527120113372803,
      "epoch": 0.28232542599398597,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004995542925021455,
      "loss": 5.5406,
      "mean_token_accuracy": 0.1637692406773567,
      "num_tokens": 6224753.0,
      "step": 3380
    },
    {
      "entropy": 5.791276264190674,
      "epoch": 0.28274306715669895,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004995524176476277,
      "loss": 5.7233,
      "mean_token_accuracy": 0.16027000695466995,
      "num_tokens": 6233842.0,
      "step": 3385
    },
    {
      "entropy": 5.6728884220123295,
      "epoch": 0.283160708319412,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004995505388620497,
      "loss": 5.5529,
      "mean_token_accuracy": 0.1680996060371399,
      "num_tokens": 6243467.0,
      "step": 3390
    },
    {
      "entropy": 5.4875246524810795,
      "epoch": 0.28357834948212496,
      "grad_norm": 1.125,
      "learning_rate": 0.000499548656145444,
      "loss": 5.592,
      "mean_token_accuracy": 0.1577073633670807,
      "num_tokens": 6253263.0,
      "step": 3395
    },
    {
      "entropy": 5.601888036727905,
      "epoch": 0.28399599064483794,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004995467694978439,
      "loss": 5.4437,
      "mean_token_accuracy": 0.17447730600833894,
      "num_tokens": 6262249.0,
      "step": 3400
    },
    {
      "entropy": 5.494926261901855,
      "epoch": 0.28441363180755097,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004995448789192823,
      "loss": 5.4555,
      "mean_token_accuracy": 0.17069571316242219,
      "num_tokens": 6271154.0,
      "step": 3405
    },
    {
      "entropy": 5.508152675628662,
      "epoch": 0.28483127297026395,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004995429844097922,
      "loss": 5.5272,
      "mean_token_accuracy": 0.174279648065567,
      "num_tokens": 6280377.0,
      "step": 3410
    },
    {
      "entropy": 5.585491943359375,
      "epoch": 0.2852489141329769,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004995410859694071,
      "loss": 5.5373,
      "mean_token_accuracy": 0.1684752508997917,
      "num_tokens": 6289941.0,
      "step": 3415
    },
    {
      "entropy": 5.544521903991699,
      "epoch": 0.28566655529568996,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004995391835981599,
      "loss": 5.5944,
      "mean_token_accuracy": 0.16231686770915985,
      "num_tokens": 6299281.0,
      "step": 3420
    },
    {
      "entropy": 5.651609754562378,
      "epoch": 0.28608419645840294,
      "grad_norm": 1.0625,
      "learning_rate": 0.000499537277296084,
      "loss": 5.5407,
      "mean_token_accuracy": 0.1607593074440956,
      "num_tokens": 6309408.0,
      "step": 3425
    },
    {
      "entropy": 5.559070634841919,
      "epoch": 0.2865018376211159,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004995353670632127,
      "loss": 5.5428,
      "mean_token_accuracy": 0.16397647708654403,
      "num_tokens": 6318979.0,
      "step": 3430
    },
    {
      "entropy": 5.495109891891479,
      "epoch": 0.28691947878382895,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004995334528995797,
      "loss": 5.3831,
      "mean_token_accuracy": 0.18039173036813735,
      "num_tokens": 6327616.0,
      "step": 3435
    },
    {
      "entropy": 5.461599111557007,
      "epoch": 0.28733711994654193,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004995315348052182,
      "loss": 5.4144,
      "mean_token_accuracy": 0.17610692828893662,
      "num_tokens": 6336486.0,
      "step": 3440
    },
    {
      "entropy": 5.5086901664733885,
      "epoch": 0.2877547611092549,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000499529612780162,
      "loss": 5.4734,
      "mean_token_accuracy": 0.16618863493204117,
      "num_tokens": 6345671.0,
      "step": 3445
    },
    {
      "entropy": 5.6285436153411865,
      "epoch": 0.28817240227196794,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004995276868244446,
      "loss": 5.5721,
      "mean_token_accuracy": 0.16796167194843292,
      "num_tokens": 6354534.0,
      "step": 3450
    },
    {
      "entropy": 5.510032081604004,
      "epoch": 0.2885900434346809,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004995257569380998,
      "loss": 5.4158,
      "mean_token_accuracy": 0.17104145437479018,
      "num_tokens": 6364052.0,
      "step": 3455
    },
    {
      "entropy": 5.539919042587281,
      "epoch": 0.2890076845973939,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004995238231211614,
      "loss": 5.5457,
      "mean_token_accuracy": 0.1703997939825058,
      "num_tokens": 6373305.0,
      "step": 3460
    },
    {
      "entropy": 5.5487676620483395,
      "epoch": 0.28942532576010693,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004995218853736631,
      "loss": 5.4752,
      "mean_token_accuracy": 0.16317730098962785,
      "num_tokens": 6382673.0,
      "step": 3465
    },
    {
      "entropy": 5.597512865066529,
      "epoch": 0.2898429669228199,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004995199436956391,
      "loss": 5.5808,
      "mean_token_accuracy": 0.16054316014051437,
      "num_tokens": 6392122.0,
      "step": 3470
    },
    {
      "entropy": 5.643544769287109,
      "epoch": 0.2902606080855329,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004995179980871231,
      "loss": 5.5854,
      "mean_token_accuracy": 0.16072637736797332,
      "num_tokens": 6401130.0,
      "step": 3475
    },
    {
      "entropy": 5.538010168075561,
      "epoch": 0.2906782492482459,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004995160485481493,
      "loss": 5.4488,
      "mean_token_accuracy": 0.16509442627429963,
      "num_tokens": 6409471.0,
      "step": 3480
    },
    {
      "entropy": 5.478931093215943,
      "epoch": 0.2910958904109589,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000499514095078752,
      "loss": 5.479,
      "mean_token_accuracy": 0.16667304188013077,
      "num_tokens": 6418902.0,
      "step": 3485
    },
    {
      "entropy": 5.574909925460815,
      "epoch": 0.2915135315736719,
      "grad_norm": 1.03125,
      "learning_rate": 0.000499512137678965,
      "loss": 5.5514,
      "mean_token_accuracy": 0.16285862326622008,
      "num_tokens": 6428529.0,
      "step": 3490
    },
    {
      "entropy": 5.5814121723175045,
      "epoch": 0.2919311727363849,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004995101763488229,
      "loss": 5.4113,
      "mean_token_accuracy": 0.1701946422457695,
      "num_tokens": 6437460.0,
      "step": 3495
    },
    {
      "entropy": 5.531909990310669,
      "epoch": 0.2923488138990979,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004995082110883599,
      "loss": 5.4605,
      "mean_token_accuracy": 0.17182532101869583,
      "num_tokens": 6446005.0,
      "step": 3500
    },
    {
      "entropy": 5.5357420444488525,
      "epoch": 0.29276645506181087,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004995062418976104,
      "loss": 5.5271,
      "mean_token_accuracy": 0.16687075346708297,
      "num_tokens": 6455241.0,
      "step": 3505
    },
    {
      "entropy": 5.548692512512207,
      "epoch": 0.2931840962245239,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004995042687766089,
      "loss": 5.5018,
      "mean_token_accuracy": 0.1705710157752037,
      "num_tokens": 6464748.0,
      "step": 3510
    },
    {
      "entropy": 5.615511178970337,
      "epoch": 0.2936017373872369,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004995022917253899,
      "loss": 5.5552,
      "mean_token_accuracy": 0.1701499566435814,
      "num_tokens": 6473834.0,
      "step": 3515
    },
    {
      "entropy": 5.533133888244629,
      "epoch": 0.29401937854994986,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004995003107439882,
      "loss": 5.4936,
      "mean_token_accuracy": 0.16701297461986542,
      "num_tokens": 6482169.0,
      "step": 3520
    },
    {
      "entropy": 5.459322071075439,
      "epoch": 0.2944370197126629,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004994983258324383,
      "loss": 5.4392,
      "mean_token_accuracy": 0.1680704489350319,
      "num_tokens": 6491798.0,
      "step": 3525
    },
    {
      "entropy": 5.4757795333862305,
      "epoch": 0.2948546608753759,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004994963369907748,
      "loss": 5.374,
      "mean_token_accuracy": 0.16915201246738434,
      "num_tokens": 6500143.0,
      "step": 3530
    },
    {
      "entropy": 5.4345615863800045,
      "epoch": 0.29527230203808885,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004994943442190328,
      "loss": 5.5127,
      "mean_token_accuracy": 0.16544250398874283,
      "num_tokens": 6509745.0,
      "step": 3535
    },
    {
      "entropy": 5.650967645645141,
      "epoch": 0.2956899432008019,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004994923475172471,
      "loss": 5.5848,
      "mean_token_accuracy": 0.15887098908424377,
      "num_tokens": 6520931.0,
      "step": 3540
    },
    {
      "entropy": 5.5606303215026855,
      "epoch": 0.29610758436351486,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004994903468854527,
      "loss": 5.5197,
      "mean_token_accuracy": 0.16634719371795653,
      "num_tokens": 6529774.0,
      "step": 3545
    },
    {
      "entropy": 5.544323110580445,
      "epoch": 0.29652522552622784,
      "grad_norm": 1.125,
      "learning_rate": 0.0004994883423236843,
      "loss": 5.4345,
      "mean_token_accuracy": 0.16737744957208633,
      "num_tokens": 6538922.0,
      "step": 3550
    },
    {
      "entropy": 5.524617290496826,
      "epoch": 0.2969428666889409,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004994863338319775,
      "loss": 5.4913,
      "mean_token_accuracy": 0.17015527039766312,
      "num_tokens": 6548891.0,
      "step": 3555
    },
    {
      "entropy": 5.405666828155518,
      "epoch": 0.29736050785165385,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004994843214103672,
      "loss": 5.306,
      "mean_token_accuracy": 0.17207494229078293,
      "num_tokens": 6557025.0,
      "step": 3560
    },
    {
      "entropy": 5.56685996055603,
      "epoch": 0.29777814901436683,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004994823050588886,
      "loss": 5.4925,
      "mean_token_accuracy": 0.16344626247882843,
      "num_tokens": 6567037.0,
      "step": 3565
    },
    {
      "entropy": 5.533381080627441,
      "epoch": 0.29819579017707987,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000499480284777577,
      "loss": 5.5141,
      "mean_token_accuracy": 0.1677533283829689,
      "num_tokens": 6576766.0,
      "step": 3570
    },
    {
      "entropy": 5.5538324356079105,
      "epoch": 0.29861343133979285,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004994782605664677,
      "loss": 5.4885,
      "mean_token_accuracy": 0.16958095133304596,
      "num_tokens": 6585576.0,
      "step": 3575
    },
    {
      "entropy": 5.584893608093262,
      "epoch": 0.2990310725025058,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004994762324255965,
      "loss": 5.559,
      "mean_token_accuracy": 0.16368599086999894,
      "num_tokens": 6594463.0,
      "step": 3580
    },
    {
      "entropy": 5.512655591964721,
      "epoch": 0.29944871366521886,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004994742003549986,
      "loss": 5.4237,
      "mean_token_accuracy": 0.16809990853071213,
      "num_tokens": 6603651.0,
      "step": 3585
    },
    {
      "entropy": 5.561243295669556,
      "epoch": 0.29986635482793184,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004994721643547095,
      "loss": 5.503,
      "mean_token_accuracy": 0.16976088881492615,
      "num_tokens": 6612950.0,
      "step": 3590
    },
    {
      "entropy": 5.553755426406861,
      "epoch": 0.3002839959906448,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000499470124424765,
      "loss": 5.44,
      "mean_token_accuracy": 0.17298568338155745,
      "num_tokens": 6622013.0,
      "step": 3595
    },
    {
      "entropy": 5.548866319656372,
      "epoch": 0.30070163715335785,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004994680805652008,
      "loss": 5.5163,
      "mean_token_accuracy": 0.16933971494436265,
      "num_tokens": 6632046.0,
      "step": 3600
    },
    {
      "entropy": 5.511341762542725,
      "epoch": 0.3011192783160708,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004994660327760527,
      "loss": 5.5702,
      "mean_token_accuracy": 0.17278106957674028,
      "num_tokens": 6641169.0,
      "step": 3605
    },
    {
      "entropy": 5.510961246490479,
      "epoch": 0.3015369194787838,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004994639810573565,
      "loss": 5.3725,
      "mean_token_accuracy": 0.17695365101099014,
      "num_tokens": 6649663.0,
      "step": 3610
    },
    {
      "entropy": 5.623845148086548,
      "epoch": 0.30195456064149684,
      "grad_norm": 1.140625,
      "learning_rate": 0.000499461925409148,
      "loss": 5.5665,
      "mean_token_accuracy": 0.16110052913427353,
      "num_tokens": 6659448.0,
      "step": 3615
    },
    {
      "entropy": 5.590248870849609,
      "epoch": 0.3023722018042098,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004994598658314634,
      "loss": 5.5246,
      "mean_token_accuracy": 0.16626361906528472,
      "num_tokens": 6670314.0,
      "step": 3620
    },
    {
      "entropy": 5.511123561859131,
      "epoch": 0.3027898429669228,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004994578023243386,
      "loss": 5.4871,
      "mean_token_accuracy": 0.1732531487941742,
      "num_tokens": 6679413.0,
      "step": 3625
    },
    {
      "entropy": 5.49867696762085,
      "epoch": 0.30320748412963583,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004994557348878098,
      "loss": 5.4627,
      "mean_token_accuracy": 0.17419134676456452,
      "num_tokens": 6688439.0,
      "step": 3630
    },
    {
      "entropy": 5.5494426727294925,
      "epoch": 0.3036251252923488,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004994536635219133,
      "loss": 5.4403,
      "mean_token_accuracy": 0.17038834542036058,
      "num_tokens": 6697471.0,
      "step": 3635
    },
    {
      "entropy": 5.586328601837158,
      "epoch": 0.3040427664550618,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004994515882266852,
      "loss": 5.4932,
      "mean_token_accuracy": 0.17040248662233354,
      "num_tokens": 6706698.0,
      "step": 3640
    },
    {
      "entropy": 5.415295648574829,
      "epoch": 0.3044604076177748,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004994495090021618,
      "loss": 5.4289,
      "mean_token_accuracy": 0.17661873996257782,
      "num_tokens": 6715676.0,
      "step": 3645
    },
    {
      "entropy": 5.427714014053345,
      "epoch": 0.3048780487804878,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004994474258483796,
      "loss": 5.4447,
      "mean_token_accuracy": 0.17413119971752167,
      "num_tokens": 6723818.0,
      "step": 3650
    },
    {
      "entropy": 5.544740056991577,
      "epoch": 0.3052956899432008,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004994453387653751,
      "loss": 5.4632,
      "mean_token_accuracy": 0.1687764897942543,
      "num_tokens": 6733450.0,
      "step": 3655
    },
    {
      "entropy": 5.481561374664307,
      "epoch": 0.3057133311059138,
      "grad_norm": 1.125,
      "learning_rate": 0.0004994432477531847,
      "loss": 5.4078,
      "mean_token_accuracy": 0.17082249820232392,
      "num_tokens": 6743142.0,
      "step": 3660
    },
    {
      "entropy": 5.436155891418457,
      "epoch": 0.3061309722686268,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004994411528118452,
      "loss": 5.3768,
      "mean_token_accuracy": 0.17255899906158448,
      "num_tokens": 6752185.0,
      "step": 3665
    },
    {
      "entropy": 5.477898073196411,
      "epoch": 0.30654861343133977,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004994390539413931,
      "loss": 5.4623,
      "mean_token_accuracy": 0.17468517869710923,
      "num_tokens": 6761549.0,
      "step": 3670
    },
    {
      "entropy": 5.549196720123291,
      "epoch": 0.3069662545940528,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004994369511418652,
      "loss": 5.4578,
      "mean_token_accuracy": 0.17020693123340608,
      "num_tokens": 6770395.0,
      "step": 3675
    },
    {
      "entropy": 5.6494293212890625,
      "epoch": 0.3073838957567658,
      "grad_norm": 1.25,
      "learning_rate": 0.0004994348444132983,
      "loss": 5.5408,
      "mean_token_accuracy": 0.17074840813875197,
      "num_tokens": 6779154.0,
      "step": 3680
    },
    {
      "entropy": 5.391846179962158,
      "epoch": 0.30780153691947876,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004994327337557293,
      "loss": 5.4393,
      "mean_token_accuracy": 0.17821680009365082,
      "num_tokens": 6789263.0,
      "step": 3685
    },
    {
      "entropy": 5.5761597633361815,
      "epoch": 0.3082191780821918,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004994306191691952,
      "loss": 5.5237,
      "mean_token_accuracy": 0.16877364367246628,
      "num_tokens": 6797985.0,
      "step": 3690
    },
    {
      "entropy": 5.486738443374634,
      "epoch": 0.30863681924490477,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004994285006537329,
      "loss": 5.4226,
      "mean_token_accuracy": 0.1722109720110893,
      "num_tokens": 6807566.0,
      "step": 3695
    },
    {
      "entropy": 5.487214660644531,
      "epoch": 0.30905446040761775,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004994263782093796,
      "loss": 5.461,
      "mean_token_accuracy": 0.17012524008750915,
      "num_tokens": 6815637.0,
      "step": 3700
    },
    {
      "entropy": 5.505376195907592,
      "epoch": 0.3094721015703308,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004994242518361723,
      "loss": 5.3487,
      "mean_token_accuracy": 0.17713512480258942,
      "num_tokens": 6823721.0,
      "step": 3705
    },
    {
      "entropy": 5.575809669494629,
      "epoch": 0.30988974273304376,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004994221215341484,
      "loss": 5.5168,
      "mean_token_accuracy": 0.17037255465984344,
      "num_tokens": 6833437.0,
      "step": 3710
    },
    {
      "entropy": 5.5108507633209225,
      "epoch": 0.31030738389575674,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004994199873033452,
      "loss": 5.5761,
      "mean_token_accuracy": 0.171338027715683,
      "num_tokens": 6842287.0,
      "step": 3715
    },
    {
      "entropy": 5.590300798416138,
      "epoch": 0.3107250250584698,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004994178491437999,
      "loss": 5.4002,
      "mean_token_accuracy": 0.16826316267251967,
      "num_tokens": 6851520.0,
      "step": 3720
    },
    {
      "entropy": 5.475397777557373,
      "epoch": 0.31114266622118275,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004994157070555501,
      "loss": 5.4696,
      "mean_token_accuracy": 0.17000560760498046,
      "num_tokens": 6861219.0,
      "step": 3725
    },
    {
      "entropy": 5.603476238250733,
      "epoch": 0.31156030738389573,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004994135610386331,
      "loss": 5.603,
      "mean_token_accuracy": 0.16594088822603226,
      "num_tokens": 6870525.0,
      "step": 3730
    },
    {
      "entropy": 5.626299047470093,
      "epoch": 0.31197794854660876,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004994114110930867,
      "loss": 5.52,
      "mean_token_accuracy": 0.16624996811151505,
      "num_tokens": 6880183.0,
      "step": 3735
    },
    {
      "entropy": 5.512535762786865,
      "epoch": 0.31239558970932174,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004994092572189483,
      "loss": 5.5473,
      "mean_token_accuracy": 0.1618015468120575,
      "num_tokens": 6890442.0,
      "step": 3740
    },
    {
      "entropy": 5.511465311050415,
      "epoch": 0.3128132308720348,
      "grad_norm": 1.125,
      "learning_rate": 0.0004994070994162558,
      "loss": 5.455,
      "mean_token_accuracy": 0.17312270253896714,
      "num_tokens": 6900091.0,
      "step": 3745
    },
    {
      "entropy": 5.496584510803222,
      "epoch": 0.31323087203474775,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004994049376850471,
      "loss": 5.499,
      "mean_token_accuracy": 0.16660811454057695,
      "num_tokens": 6910489.0,
      "step": 3750
    },
    {
      "entropy": 5.597193670272827,
      "epoch": 0.31364851319746073,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004994027720253596,
      "loss": 5.4953,
      "mean_token_accuracy": 0.16707699447870256,
      "num_tokens": 6919895.0,
      "step": 3755
    },
    {
      "entropy": 5.6900543689727785,
      "epoch": 0.31406615436017377,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004994006024372315,
      "loss": 5.6114,
      "mean_token_accuracy": 0.16193894147872925,
      "num_tokens": 6929511.0,
      "step": 3760
    },
    {
      "entropy": 5.477146863937378,
      "epoch": 0.31448379552288674,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004993984289207007,
      "loss": 5.388,
      "mean_token_accuracy": 0.1762948825955391,
      "num_tokens": 6938059.0,
      "step": 3765
    },
    {
      "entropy": 5.497006607055664,
      "epoch": 0.3149014366855997,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004993962514758054,
      "loss": 5.4172,
      "mean_token_accuracy": 0.1702394813299179,
      "num_tokens": 6946894.0,
      "step": 3770
    },
    {
      "entropy": 5.554160165786743,
      "epoch": 0.31531907784831276,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004993940701025836,
      "loss": 5.4148,
      "mean_token_accuracy": 0.16560422331094743,
      "num_tokens": 6955494.0,
      "step": 3775
    },
    {
      "entropy": 5.455316400527954,
      "epoch": 0.31573671901102573,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004993918848010734,
      "loss": 5.5254,
      "mean_token_accuracy": 0.16878335326910018,
      "num_tokens": 6964347.0,
      "step": 3780
    },
    {
      "entropy": 5.533470582962036,
      "epoch": 0.3161543601737387,
      "grad_norm": 1.125,
      "learning_rate": 0.0004993896955713133,
      "loss": 5.4786,
      "mean_token_accuracy": 0.1637830153107643,
      "num_tokens": 6974750.0,
      "step": 3785
    },
    {
      "entropy": 5.582107496261597,
      "epoch": 0.31657200133645175,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004993875024133414,
      "loss": 5.4921,
      "mean_token_accuracy": 0.17069665789604188,
      "num_tokens": 6984228.0,
      "step": 3790
    },
    {
      "entropy": 5.4917913436889645,
      "epoch": 0.3169896424991647,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004993853053271962,
      "loss": 5.4694,
      "mean_token_accuracy": 0.16878245770931244,
      "num_tokens": 6993173.0,
      "step": 3795
    },
    {
      "entropy": 5.706063556671142,
      "epoch": 0.3174072836618777,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004993831043129161,
      "loss": 5.6669,
      "mean_token_accuracy": 0.1557183623313904,
      "num_tokens": 7003486.0,
      "step": 3800
    },
    {
      "entropy": 5.500661993026734,
      "epoch": 0.31782492482459074,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004993808993705397,
      "loss": 5.4256,
      "mean_token_accuracy": 0.1763479545712471,
      "num_tokens": 7011985.0,
      "step": 3805
    },
    {
      "entropy": 5.481622362136841,
      "epoch": 0.3182425659873037,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004993786905001055,
      "loss": 5.3832,
      "mean_token_accuracy": 0.17533033937215806,
      "num_tokens": 7021148.0,
      "step": 3810
    },
    {
      "entropy": 5.464303827285766,
      "epoch": 0.3186602071500167,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004993764777016524,
      "loss": 5.4438,
      "mean_token_accuracy": 0.16824772357940673,
      "num_tokens": 7030110.0,
      "step": 3815
    },
    {
      "entropy": 5.539179086685181,
      "epoch": 0.31907784831272973,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004993742609752188,
      "loss": 5.5134,
      "mean_token_accuracy": 0.17360101789236068,
      "num_tokens": 7039428.0,
      "step": 3820
    },
    {
      "entropy": 5.614913988113403,
      "epoch": 0.3194954894754427,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004993720403208438,
      "loss": 5.5701,
      "mean_token_accuracy": 0.16541504263877868,
      "num_tokens": 7048703.0,
      "step": 3825
    },
    {
      "entropy": 5.552296113967896,
      "epoch": 0.3199131306381557,
      "grad_norm": 1.21875,
      "learning_rate": 0.000499369815738566,
      "loss": 5.4614,
      "mean_token_accuracy": 0.1668523982167244,
      "num_tokens": 7057612.0,
      "step": 3830
    },
    {
      "entropy": 5.5432675838470455,
      "epoch": 0.3203307718008687,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004993675872284248,
      "loss": 5.416,
      "mean_token_accuracy": 0.17318572700023652,
      "num_tokens": 7066660.0,
      "step": 3835
    },
    {
      "entropy": 5.433846664428711,
      "epoch": 0.3207484129635817,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004993653547904586,
      "loss": 5.3986,
      "mean_token_accuracy": 0.16264898627996444,
      "num_tokens": 7075368.0,
      "step": 3840
    },
    {
      "entropy": 5.536958503723144,
      "epoch": 0.3211660541262947,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000499363118424707,
      "loss": 5.5073,
      "mean_token_accuracy": 0.17436618506908416,
      "num_tokens": 7085276.0,
      "step": 3845
    },
    {
      "entropy": 5.533533716201783,
      "epoch": 0.3215836952890077,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004993608781312088,
      "loss": 5.4679,
      "mean_token_accuracy": 0.16549077928066253,
      "num_tokens": 7094292.0,
      "step": 3850
    },
    {
      "entropy": 5.648198843002319,
      "epoch": 0.3220013364517207,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004993586339100035,
      "loss": 5.676,
      "mean_token_accuracy": 0.1579186387360096,
      "num_tokens": 7104712.0,
      "step": 3855
    },
    {
      "entropy": 5.6159360885620115,
      "epoch": 0.32241897761443367,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004993563857611301,
      "loss": 5.4774,
      "mean_token_accuracy": 0.17931303083896638,
      "num_tokens": 7113297.0,
      "step": 3860
    },
    {
      "entropy": 5.473030376434326,
      "epoch": 0.3228366187771467,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004993541336846283,
      "loss": 5.4821,
      "mean_token_accuracy": 0.16101839244365693,
      "num_tokens": 7122706.0,
      "step": 3865
    },
    {
      "entropy": 5.521943998336792,
      "epoch": 0.3232542599398597,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004993518776805372,
      "loss": 5.4584,
      "mean_token_accuracy": 0.1689613327383995,
      "num_tokens": 7131939.0,
      "step": 3870
    },
    {
      "entropy": 5.393316602706909,
      "epoch": 0.32367190110257266,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004993496177488965,
      "loss": 5.3407,
      "mean_token_accuracy": 0.1780867889523506,
      "num_tokens": 7141016.0,
      "step": 3875
    },
    {
      "entropy": 5.4151207447052006,
      "epoch": 0.3240895422652857,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004993473538897457,
      "loss": 5.2848,
      "mean_token_accuracy": 0.1836669310927391,
      "num_tokens": 7149177.0,
      "step": 3880
    },
    {
      "entropy": 5.486827421188354,
      "epoch": 0.32450718342799867,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004993450861031243,
      "loss": 5.4077,
      "mean_token_accuracy": 0.17718248218297958,
      "num_tokens": 7157839.0,
      "step": 3885
    },
    {
      "entropy": 5.62509331703186,
      "epoch": 0.32492482459071165,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004993428143890722,
      "loss": 5.5658,
      "mean_token_accuracy": 0.1595238044857979,
      "num_tokens": 7167759.0,
      "step": 3890
    },
    {
      "entropy": 5.622683382034301,
      "epoch": 0.3253424657534247,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004993405387476291,
      "loss": 5.506,
      "mean_token_accuracy": 0.16096927523612975,
      "num_tokens": 7177141.0,
      "step": 3895
    },
    {
      "entropy": 5.4179102897644045,
      "epoch": 0.32576010691613766,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004993382591788349,
      "loss": 5.4166,
      "mean_token_accuracy": 0.17410117983818055,
      "num_tokens": 7185350.0,
      "step": 3900
    },
    {
      "entropy": 5.40603928565979,
      "epoch": 0.32617774807885064,
      "grad_norm": 1.125,
      "learning_rate": 0.0004993359756827293,
      "loss": 5.3604,
      "mean_token_accuracy": 0.1728378564119339,
      "num_tokens": 7194073.0,
      "step": 3905
    },
    {
      "entropy": 5.532111930847168,
      "epoch": 0.32659538924156367,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004993336882593524,
      "loss": 5.5315,
      "mean_token_accuracy": 0.16291088610887527,
      "num_tokens": 7203421.0,
      "step": 3910
    },
    {
      "entropy": 5.449473905563354,
      "epoch": 0.32701303040427665,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004993313969087444,
      "loss": 5.2971,
      "mean_token_accuracy": 0.1715938299894333,
      "num_tokens": 7212548.0,
      "step": 3915
    },
    {
      "entropy": 5.540735912322998,
      "epoch": 0.32743067156698963,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004993291016309452,
      "loss": 5.5334,
      "mean_token_accuracy": 0.16757371574640273,
      "num_tokens": 7221668.0,
      "step": 3920
    },
    {
      "entropy": 5.507889461517334,
      "epoch": 0.32784831272970266,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000499326802425995,
      "loss": 5.4383,
      "mean_token_accuracy": 0.1652974471449852,
      "num_tokens": 7231775.0,
      "step": 3925
    },
    {
      "entropy": 5.555531215667725,
      "epoch": 0.32826595389241564,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004993244992939342,
      "loss": 5.4767,
      "mean_token_accuracy": 0.16909126937389374,
      "num_tokens": 7241264.0,
      "step": 3930
    },
    {
      "entropy": 5.477525520324707,
      "epoch": 0.3286835950551286,
      "grad_norm": 1.125,
      "learning_rate": 0.000499322192234803,
      "loss": 5.491,
      "mean_token_accuracy": 0.16701171547174454,
      "num_tokens": 7250519.0,
      "step": 3935
    },
    {
      "entropy": 5.6636481285095215,
      "epoch": 0.32910123621784165,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004993198812486418,
      "loss": 5.5395,
      "mean_token_accuracy": 0.16968121230602265,
      "num_tokens": 7259533.0,
      "step": 3940
    },
    {
      "entropy": 5.493396806716919,
      "epoch": 0.32951887738055463,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000499317566335491,
      "loss": 5.3672,
      "mean_token_accuracy": 0.17070206999778748,
      "num_tokens": 7268980.0,
      "step": 3945
    },
    {
      "entropy": 5.480592918395996,
      "epoch": 0.3299365185432676,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004993152474953913,
      "loss": 5.4578,
      "mean_token_accuracy": 0.16594725996255874,
      "num_tokens": 7277126.0,
      "step": 3950
    },
    {
      "entropy": 5.528022766113281,
      "epoch": 0.33035415970598064,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004993129247283831,
      "loss": 5.4208,
      "mean_token_accuracy": 0.17594209760427476,
      "num_tokens": 7286223.0,
      "step": 3955
    },
    {
      "entropy": 5.486289978027344,
      "epoch": 0.3307718008686936,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004993105980345071,
      "loss": 5.4408,
      "mean_token_accuracy": 0.1727459043264389,
      "num_tokens": 7295054.0,
      "step": 3960
    },
    {
      "entropy": 5.5196291446685795,
      "epoch": 0.3311894420314066,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004993082674138042,
      "loss": 5.4602,
      "mean_token_accuracy": 0.16758015602827073,
      "num_tokens": 7304502.0,
      "step": 3965
    },
    {
      "entropy": 5.555356788635254,
      "epoch": 0.33160708319411963,
      "grad_norm": 1.015625,
      "learning_rate": 0.000499305932866315,
      "loss": 5.4341,
      "mean_token_accuracy": 0.17165492922067643,
      "num_tokens": 7314093.0,
      "step": 3970
    },
    {
      "entropy": 5.50099024772644,
      "epoch": 0.3320247243568326,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004993035943920805,
      "loss": 5.4572,
      "mean_token_accuracy": 0.1696681097149849,
      "num_tokens": 7323059.0,
      "step": 3975
    },
    {
      "entropy": 5.6192985534667965,
      "epoch": 0.3324423655195456,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004993012519911416,
      "loss": 5.5943,
      "mean_token_accuracy": 0.16187372356653212,
      "num_tokens": 7332596.0,
      "step": 3980
    },
    {
      "entropy": 5.532523775100708,
      "epoch": 0.3328600066822586,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004992989056635392,
      "loss": 5.3946,
      "mean_token_accuracy": 0.17754316031932832,
      "num_tokens": 7341671.0,
      "step": 3985
    },
    {
      "entropy": 5.4355028629302975,
      "epoch": 0.3332776478449716,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004992965554093146,
      "loss": 5.5201,
      "mean_token_accuracy": 0.16645139306783677,
      "num_tokens": 7350344.0,
      "step": 3990
    },
    {
      "entropy": 5.602485799789429,
      "epoch": 0.3336952890076846,
      "grad_norm": 1.125,
      "learning_rate": 0.0004992942012285086,
      "loss": 5.5326,
      "mean_token_accuracy": 0.1713624395430088,
      "num_tokens": 7359844.0,
      "step": 3995
    },
    {
      "entropy": 5.554539442062378,
      "epoch": 0.3341129301703976,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004992918431211629,
      "loss": 5.459,
      "mean_token_accuracy": 0.17449632287025452,
      "num_tokens": 7369377.0,
      "step": 4000
    },
    {
      "entropy": 5.482081747055053,
      "epoch": 0.3345305713331106,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004992894810873183,
      "loss": 5.4218,
      "mean_token_accuracy": 0.1696226865053177,
      "num_tokens": 7378814.0,
      "step": 4005
    },
    {
      "entropy": 5.451307106018066,
      "epoch": 0.33494821249582357,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004992871151270165,
      "loss": 5.4135,
      "mean_token_accuracy": 0.17543912082910537,
      "num_tokens": 7387333.0,
      "step": 4010
    },
    {
      "entropy": 5.422832012176514,
      "epoch": 0.3353658536585366,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004992847452402986,
      "loss": 5.4531,
      "mean_token_accuracy": 0.16946215331554412,
      "num_tokens": 7396895.0,
      "step": 4015
    },
    {
      "entropy": 5.565013027191162,
      "epoch": 0.3357834948212496,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004992823714272064,
      "loss": 5.4491,
      "mean_token_accuracy": 0.17337892353534698,
      "num_tokens": 7406430.0,
      "step": 4020
    },
    {
      "entropy": 5.513153409957885,
      "epoch": 0.33620113598396256,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004992799936877813,
      "loss": 5.3104,
      "mean_token_accuracy": 0.1838814213871956,
      "num_tokens": 7415224.0,
      "step": 4025
    },
    {
      "entropy": 5.508871841430664,
      "epoch": 0.3366187771466756,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004992776120220649,
      "loss": 5.6195,
      "mean_token_accuracy": 0.15561454445123674,
      "num_tokens": 7424250.0,
      "step": 4030
    },
    {
      "entropy": 5.605917835235596,
      "epoch": 0.3370364183093886,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004992752264300989,
      "loss": 5.4484,
      "mean_token_accuracy": 0.1740364298224449,
      "num_tokens": 7433579.0,
      "step": 4035
    },
    {
      "entropy": 5.412272262573242,
      "epoch": 0.33745405947210155,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004992728369119252,
      "loss": 5.4196,
      "mean_token_accuracy": 0.17319910526275634,
      "num_tokens": 7442621.0,
      "step": 4040
    },
    {
      "entropy": 5.564623546600342,
      "epoch": 0.3378717006348146,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004992704434675855,
      "loss": 5.5085,
      "mean_token_accuracy": 0.16694118827581406,
      "num_tokens": 7452077.0,
      "step": 4045
    },
    {
      "entropy": 5.461846494674683,
      "epoch": 0.33828934179752757,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004992680460971218,
      "loss": 5.4233,
      "mean_token_accuracy": 0.17070491313934327,
      "num_tokens": 7461146.0,
      "step": 4050
    },
    {
      "entropy": 5.454490232467651,
      "epoch": 0.33870698296024054,
      "grad_norm": 1.125,
      "learning_rate": 0.0004992656448005758,
      "loss": 5.4145,
      "mean_token_accuracy": 0.1729734495282173,
      "num_tokens": 7470335.0,
      "step": 4055
    },
    {
      "entropy": 5.581266927719116,
      "epoch": 0.3391246241229536,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004992632395779899,
      "loss": 5.4225,
      "mean_token_accuracy": 0.16888627856969834,
      "num_tokens": 7479458.0,
      "step": 4060
    },
    {
      "entropy": 5.372164392471314,
      "epoch": 0.33954226528566656,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004992608304294061,
      "loss": 5.3918,
      "mean_token_accuracy": 0.18086428940296173,
      "num_tokens": 7488005.0,
      "step": 4065
    },
    {
      "entropy": 5.390970087051391,
      "epoch": 0.33995990644837953,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004992584173548664,
      "loss": 5.3862,
      "mean_token_accuracy": 0.17391041666269302,
      "num_tokens": 7496807.0,
      "step": 4070
    },
    {
      "entropy": 5.517978620529175,
      "epoch": 0.34037754761109257,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004992560003544131,
      "loss": 5.4157,
      "mean_token_accuracy": 0.17917752861976624,
      "num_tokens": 7505383.0,
      "step": 4075
    },
    {
      "entropy": 5.567019414901734,
      "epoch": 0.34079518877380555,
      "grad_norm": 1.125,
      "learning_rate": 0.0004992535794280887,
      "loss": 5.4743,
      "mean_token_accuracy": 0.17009843438863753,
      "num_tokens": 7515093.0,
      "step": 4080
    },
    {
      "entropy": 5.514645767211914,
      "epoch": 0.3412128299365185,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004992511545759354,
      "loss": 5.4493,
      "mean_token_accuracy": 0.16668404191732406,
      "num_tokens": 7524067.0,
      "step": 4085
    },
    {
      "entropy": 5.358101844787598,
      "epoch": 0.34163047109923156,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004992487257979958,
      "loss": 5.3837,
      "mean_token_accuracy": 0.16984069496393203,
      "num_tokens": 7533197.0,
      "step": 4090
    },
    {
      "entropy": 5.556244897842407,
      "epoch": 0.34204811226194454,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004992462930943123,
      "loss": 5.4006,
      "mean_token_accuracy": 0.16619593650102615,
      "num_tokens": 7543030.0,
      "step": 4095
    },
    {
      "entropy": 5.518676137924194,
      "epoch": 0.3424657534246575,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004992438564649275,
      "loss": 5.4656,
      "mean_token_accuracy": 0.16660134345293046,
      "num_tokens": 7551556.0,
      "step": 4100
    },
    {
      "entropy": 5.500174808502197,
      "epoch": 0.34288339458737055,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004992414159098841,
      "loss": 5.3879,
      "mean_token_accuracy": 0.17086076587438584,
      "num_tokens": 7559284.0,
      "step": 4105
    },
    {
      "entropy": 5.574701356887817,
      "epoch": 0.3433010357500835,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004992389714292247,
      "loss": 5.4981,
      "mean_token_accuracy": 0.17423533499240876,
      "num_tokens": 7568750.0,
      "step": 4110
    },
    {
      "entropy": 5.394807052612305,
      "epoch": 0.3437186769127965,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004992365230229922,
      "loss": 5.358,
      "mean_token_accuracy": 0.17588887065649034,
      "num_tokens": 7578309.0,
      "step": 4115
    },
    {
      "entropy": 5.394585037231446,
      "epoch": 0.34413631807550954,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004992340706912295,
      "loss": 5.3881,
      "mean_token_accuracy": 0.17258904576301576,
      "num_tokens": 7586760.0,
      "step": 4120
    },
    {
      "entropy": 5.459982442855835,
      "epoch": 0.3445539592382225,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004992316144339794,
      "loss": 5.3502,
      "mean_token_accuracy": 0.17693475484848023,
      "num_tokens": 7595472.0,
      "step": 4125
    },
    {
      "entropy": 5.509391355514526,
      "epoch": 0.3449716004009355,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004992291542512852,
      "loss": 5.5473,
      "mean_token_accuracy": 0.15762064158916472,
      "num_tokens": 7605663.0,
      "step": 4130
    },
    {
      "entropy": 5.513069915771484,
      "epoch": 0.34538924156364853,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004992266901431895,
      "loss": 5.4359,
      "mean_token_accuracy": 0.16975632905960084,
      "num_tokens": 7616110.0,
      "step": 4135
    },
    {
      "entropy": 5.425395250320435,
      "epoch": 0.3458068827263615,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004992242221097358,
      "loss": 5.3798,
      "mean_token_accuracy": 0.17331961393356324,
      "num_tokens": 7625311.0,
      "step": 4140
    },
    {
      "entropy": 5.4703614711761475,
      "epoch": 0.3462245238890745,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004992217501509672,
      "loss": 5.4263,
      "mean_token_accuracy": 0.16921253502368927,
      "num_tokens": 7634562.0,
      "step": 4145
    },
    {
      "entropy": 5.426385354995728,
      "epoch": 0.3466421650517875,
      "grad_norm": 1.078125,
      "learning_rate": 0.000499219274266927,
      "loss": 5.3567,
      "mean_token_accuracy": 0.1739166870713234,
      "num_tokens": 7643231.0,
      "step": 4150
    },
    {
      "entropy": 5.488095760345459,
      "epoch": 0.3470598062145005,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004992167944576584,
      "loss": 5.504,
      "mean_token_accuracy": 0.1696464702486992,
      "num_tokens": 7653556.0,
      "step": 4155
    },
    {
      "entropy": 5.533686256408691,
      "epoch": 0.3474774473772135,
      "grad_norm": 1.09375,
      "learning_rate": 0.000499214310723205,
      "loss": 5.392,
      "mean_token_accuracy": 0.17667534947395325,
      "num_tokens": 7662295.0,
      "step": 4160
    },
    {
      "entropy": 5.381580638885498,
      "epoch": 0.3478950885399265,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004992118230636102,
      "loss": 5.4204,
      "mean_token_accuracy": 0.1728316977620125,
      "num_tokens": 7671429.0,
      "step": 4165
    },
    {
      "entropy": 5.365633487701416,
      "epoch": 0.3483127297026395,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004992093314789176,
      "loss": 5.3271,
      "mean_token_accuracy": 0.17085648328065872,
      "num_tokens": 7680589.0,
      "step": 4170
    },
    {
      "entropy": 5.469932985305786,
      "epoch": 0.34873037086535247,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004992068359691707,
      "loss": 5.3909,
      "mean_token_accuracy": 0.17552734017372132,
      "num_tokens": 7689005.0,
      "step": 4175
    },
    {
      "entropy": 5.3601926326751705,
      "epoch": 0.3491480120280655,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004992043365344132,
      "loss": 5.3189,
      "mean_token_accuracy": 0.179109425842762,
      "num_tokens": 7698231.0,
      "step": 4180
    },
    {
      "entropy": 5.508357238769531,
      "epoch": 0.3495656531907785,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004992018331746891,
      "loss": 5.4168,
      "mean_token_accuracy": 0.17373043149709702,
      "num_tokens": 7706678.0,
      "step": 4185
    },
    {
      "entropy": 5.527340984344482,
      "epoch": 0.34998329435349146,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004991993258900418,
      "loss": 5.4598,
      "mean_token_accuracy": 0.17227416336536408,
      "num_tokens": 7715506.0,
      "step": 4190
    },
    {
      "entropy": 5.5114830493927,
      "epoch": 0.3504009355162045,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004991968146805156,
      "loss": 5.4399,
      "mean_token_accuracy": 0.17181310206651687,
      "num_tokens": 7726549.0,
      "step": 4195
    },
    {
      "entropy": 5.487708234786988,
      "epoch": 0.35081857667891747,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004991942995461542,
      "loss": 5.5264,
      "mean_token_accuracy": 0.16997541934251786,
      "num_tokens": 7735563.0,
      "step": 4200
    },
    {
      "entropy": 5.518128681182861,
      "epoch": 0.35123621784163045,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004991917804870018,
      "loss": 5.4587,
      "mean_token_accuracy": 0.1694946900010109,
      "num_tokens": 7746144.0,
      "step": 4205
    },
    {
      "entropy": 5.485707473754883,
      "epoch": 0.3516538590043435,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004991892575031024,
      "loss": 5.465,
      "mean_token_accuracy": 0.17360012531280516,
      "num_tokens": 7755394.0,
      "step": 4210
    },
    {
      "entropy": 5.425092363357544,
      "epoch": 0.35207150016705646,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004991867305945001,
      "loss": 5.4158,
      "mean_token_accuracy": 0.17738278210163116,
      "num_tokens": 7764855.0,
      "step": 4215
    },
    {
      "entropy": 5.414138317108154,
      "epoch": 0.35248914132976944,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004991841997612394,
      "loss": 5.3436,
      "mean_token_accuracy": 0.17033103108406067,
      "num_tokens": 7773394.0,
      "step": 4220
    },
    {
      "entropy": 5.51355562210083,
      "epoch": 0.3529067824924825,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004991816650033643,
      "loss": 5.3986,
      "mean_token_accuracy": 0.1705158069729805,
      "num_tokens": 7783111.0,
      "step": 4225
    },
    {
      "entropy": 5.555590009689331,
      "epoch": 0.35332442365519545,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004991791263209194,
      "loss": 5.5372,
      "mean_token_accuracy": 0.1695098474621773,
      "num_tokens": 7792569.0,
      "step": 4230
    },
    {
      "entropy": 5.462711668014526,
      "epoch": 0.35374206481790843,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004991765837139491,
      "loss": 5.4291,
      "mean_token_accuracy": 0.17686702758073808,
      "num_tokens": 7801483.0,
      "step": 4235
    },
    {
      "entropy": 5.537383937835694,
      "epoch": 0.35415970598062146,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004991740371824977,
      "loss": 5.5412,
      "mean_token_accuracy": 0.1594116598367691,
      "num_tokens": 7812274.0,
      "step": 4240
    },
    {
      "entropy": 5.458338975906372,
      "epoch": 0.35457734714333444,
      "grad_norm": 1.09375,
      "learning_rate": 0.00049917148672661,
      "loss": 5.3586,
      "mean_token_accuracy": 0.17292539179325103,
      "num_tokens": 7821101.0,
      "step": 4245
    },
    {
      "entropy": 5.504837369918823,
      "epoch": 0.3549949883060474,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004991689323463306,
      "loss": 5.3584,
      "mean_token_accuracy": 0.17464094161987304,
      "num_tokens": 7829805.0,
      "step": 4250
    },
    {
      "entropy": 5.426690912246704,
      "epoch": 0.35541262946876045,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004991663740417042,
      "loss": 5.4505,
      "mean_token_accuracy": 0.1664499670267105,
      "num_tokens": 7839772.0,
      "step": 4255
    },
    {
      "entropy": 5.460419940948486,
      "epoch": 0.35583027063147343,
      "grad_norm": 1.125,
      "learning_rate": 0.0004991638118127757,
      "loss": 5.5321,
      "mean_token_accuracy": 0.1616857022047043,
      "num_tokens": 7850290.0,
      "step": 4260
    },
    {
      "entropy": 5.624797105789185,
      "epoch": 0.3562479117941864,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004991612456595896,
      "loss": 5.5243,
      "mean_token_accuracy": 0.16165591925382614,
      "num_tokens": 7860180.0,
      "step": 4265
    },
    {
      "entropy": 5.3619715690612795,
      "epoch": 0.35666555295689945,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004991586755821913,
      "loss": 5.3631,
      "mean_token_accuracy": 0.18035963475704192,
      "num_tokens": 7869625.0,
      "step": 4270
    },
    {
      "entropy": 5.400491714477539,
      "epoch": 0.3570831941196124,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004991561015806254,
      "loss": 5.3234,
      "mean_token_accuracy": 0.17419369220733644,
      "num_tokens": 7878671.0,
      "step": 4275
    },
    {
      "entropy": 5.5154108047485355,
      "epoch": 0.3575008352823254,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004991535236549372,
      "loss": 5.4312,
      "mean_token_accuracy": 0.1769440293312073,
      "num_tokens": 7887225.0,
      "step": 4280
    },
    {
      "entropy": 5.422553300857544,
      "epoch": 0.35791847644503844,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004991509418051718,
      "loss": 5.3329,
      "mean_token_accuracy": 0.17964915335178375,
      "num_tokens": 7895504.0,
      "step": 4285
    },
    {
      "entropy": 5.554728364944458,
      "epoch": 0.3583361176077514,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004991483560313743,
      "loss": 5.3973,
      "mean_token_accuracy": 0.17431385666131974,
      "num_tokens": 7904875.0,
      "step": 4290
    },
    {
      "entropy": 5.465934371948242,
      "epoch": 0.3587537587704644,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00049914576633359,
      "loss": 5.4618,
      "mean_token_accuracy": 0.16960846036672592,
      "num_tokens": 7914747.0,
      "step": 4295
    },
    {
      "entropy": 5.4626648902893065,
      "epoch": 0.3591713999331774,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004991431727118643,
      "loss": 5.4289,
      "mean_token_accuracy": 0.17281985133886338,
      "num_tokens": 7923858.0,
      "step": 4300
    },
    {
      "entropy": 5.457186269760132,
      "epoch": 0.3595890410958904,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004991405751662426,
      "loss": 5.3874,
      "mean_token_accuracy": 0.1731380209326744,
      "num_tokens": 7932396.0,
      "step": 4305
    },
    {
      "entropy": 5.472230291366577,
      "epoch": 0.3600066822586034,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004991379736967703,
      "loss": 5.3909,
      "mean_token_accuracy": 0.17094642221927642,
      "num_tokens": 7940763.0,
      "step": 4310
    },
    {
      "entropy": 5.391937160491944,
      "epoch": 0.3604243234213164,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004991353683034929,
      "loss": 5.3969,
      "mean_token_accuracy": 0.17015578150749205,
      "num_tokens": 7949424.0,
      "step": 4315
    },
    {
      "entropy": 5.456928300857544,
      "epoch": 0.3608419645840294,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004991327589864563,
      "loss": 5.363,
      "mean_token_accuracy": 0.17916346043348313,
      "num_tokens": 7958805.0,
      "step": 4320
    },
    {
      "entropy": 5.511008167266846,
      "epoch": 0.3612596057467424,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004991301457457057,
      "loss": 5.4862,
      "mean_token_accuracy": 0.17318558990955352,
      "num_tokens": 7967859.0,
      "step": 4325
    },
    {
      "entropy": 5.611777591705322,
      "epoch": 0.3616772469094554,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004991275285812873,
      "loss": 5.6055,
      "mean_token_accuracy": 0.16365025639533998,
      "num_tokens": 7978357.0,
      "step": 4330
    },
    {
      "entropy": 5.593785142898559,
      "epoch": 0.3620948880721684,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004991249074932466,
      "loss": 5.4584,
      "mean_token_accuracy": 0.17306540310382842,
      "num_tokens": 7986637.0,
      "step": 4335
    },
    {
      "entropy": 5.384535217285157,
      "epoch": 0.36251252923488136,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004991222824816296,
      "loss": 5.383,
      "mean_token_accuracy": 0.17511666864156722,
      "num_tokens": 7995954.0,
      "step": 4340
    },
    {
      "entropy": 5.530560398101807,
      "epoch": 0.3629301703975944,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004991196535464824,
      "loss": 5.4442,
      "mean_token_accuracy": 0.1695959284901619,
      "num_tokens": 8005193.0,
      "step": 4345
    },
    {
      "entropy": 5.577112483978271,
      "epoch": 0.3633478115603074,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004991170206878508,
      "loss": 5.4548,
      "mean_token_accuracy": 0.17263143956661225,
      "num_tokens": 8014419.0,
      "step": 4350
    },
    {
      "entropy": 5.475562572479248,
      "epoch": 0.36376545272302035,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000499114383905781,
      "loss": 5.4053,
      "mean_token_accuracy": 0.1665061816573143,
      "num_tokens": 8024133.0,
      "step": 4355
    },
    {
      "entropy": 5.413761997222901,
      "epoch": 0.3641830938857334,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004991117432003192,
      "loss": 5.3859,
      "mean_token_accuracy": 0.17992129921913147,
      "num_tokens": 8033059.0,
      "step": 4360
    },
    {
      "entropy": 5.5130657196044925,
      "epoch": 0.36460073504844637,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004991090985715116,
      "loss": 5.4517,
      "mean_token_accuracy": 0.1732624128460884,
      "num_tokens": 8041866.0,
      "step": 4365
    },
    {
      "entropy": 5.514360857009888,
      "epoch": 0.36501837621115935,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004991064500194043,
      "loss": 5.4752,
      "mean_token_accuracy": 0.16933687031269073,
      "num_tokens": 8051541.0,
      "step": 4370
    },
    {
      "entropy": 5.554482650756836,
      "epoch": 0.3654360173738724,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004991037975440439,
      "loss": 5.5428,
      "mean_token_accuracy": 0.16507816761732103,
      "num_tokens": 8061964.0,
      "step": 4375
    },
    {
      "entropy": 5.407950353622437,
      "epoch": 0.36585365853658536,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004991011411454769,
      "loss": 5.354,
      "mean_token_accuracy": 0.17312078028917313,
      "num_tokens": 8071006.0,
      "step": 4380
    },
    {
      "entropy": 5.431437063217163,
      "epoch": 0.36627129969929834,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004990984808237495,
      "loss": 5.3646,
      "mean_token_accuracy": 0.1760965332388878,
      "num_tokens": 8080104.0,
      "step": 4385
    },
    {
      "entropy": 5.521042060852051,
      "epoch": 0.36668894086201137,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004990958165789086,
      "loss": 5.4173,
      "mean_token_accuracy": 0.17236875146627426,
      "num_tokens": 8089397.0,
      "step": 4390
    },
    {
      "entropy": 5.438767433166504,
      "epoch": 0.36710658202472435,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004990931484110005,
      "loss": 5.4597,
      "mean_token_accuracy": 0.16909443438053132,
      "num_tokens": 8099055.0,
      "step": 4395
    },
    {
      "entropy": 5.532427740097046,
      "epoch": 0.3675242231874373,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004990904763200722,
      "loss": 5.3883,
      "mean_token_accuracy": 0.17435784935951232,
      "num_tokens": 8107609.0,
      "step": 4400
    },
    {
      "entropy": 5.514212989807129,
      "epoch": 0.36794186435015036,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004990878003061705,
      "loss": 5.4019,
      "mean_token_accuracy": 0.17783958315849305,
      "num_tokens": 8116905.0,
      "step": 4405
    },
    {
      "entropy": 5.405974292755127,
      "epoch": 0.36835950551286334,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004990851203693419,
      "loss": 5.4,
      "mean_token_accuracy": 0.17210870534181594,
      "num_tokens": 8125580.0,
      "step": 4410
    },
    {
      "entropy": 5.415756464004517,
      "epoch": 0.3687771466755763,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004990824365096336,
      "loss": 5.326,
      "mean_token_accuracy": 0.17104125618934632,
      "num_tokens": 8134568.0,
      "step": 4415
    },
    {
      "entropy": 5.375641012191773,
      "epoch": 0.36919478783828935,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004990797487270926,
      "loss": 5.2957,
      "mean_token_accuracy": 0.17495275139808655,
      "num_tokens": 8143044.0,
      "step": 4420
    },
    {
      "entropy": 5.435410976409912,
      "epoch": 0.36961242900100233,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004990770570217659,
      "loss": 5.4167,
      "mean_token_accuracy": 0.17094517052173613,
      "num_tokens": 8153701.0,
      "step": 4425
    },
    {
      "entropy": 5.442280197143555,
      "epoch": 0.37003007016371536,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004990743613937005,
      "loss": 5.3341,
      "mean_token_accuracy": 0.1752011701464653,
      "num_tokens": 8163264.0,
      "step": 4430
    },
    {
      "entropy": 5.401455450057983,
      "epoch": 0.37044771132642834,
      "grad_norm": 1.125,
      "learning_rate": 0.0004990716618429436,
      "loss": 5.4297,
      "mean_token_accuracy": 0.16744692623615265,
      "num_tokens": 8173332.0,
      "step": 4435
    },
    {
      "entropy": 5.488806295394897,
      "epoch": 0.3708653524891413,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004990689583695428,
      "loss": 5.4233,
      "mean_token_accuracy": 0.1716279938817024,
      "num_tokens": 8181881.0,
      "step": 4440
    },
    {
      "entropy": 5.5240256786346436,
      "epoch": 0.37128299365185435,
      "grad_norm": 1.203125,
      "learning_rate": 0.000499066250973545,
      "loss": 5.4662,
      "mean_token_accuracy": 0.16734483540058137,
      "num_tokens": 8190771.0,
      "step": 4445
    },
    {
      "entropy": 5.457997035980225,
      "epoch": 0.37170063481456733,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004990635396549978,
      "loss": 5.3942,
      "mean_token_accuracy": 0.1713161513209343,
      "num_tokens": 8200740.0,
      "step": 4450
    },
    {
      "entropy": 5.393514347076416,
      "epoch": 0.3721182759772803,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004990608244139486,
      "loss": 5.3827,
      "mean_token_accuracy": 0.16958940178155898,
      "num_tokens": 8209935.0,
      "step": 4455
    },
    {
      "entropy": 5.5158362865448,
      "epoch": 0.37253591713999334,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000499058105250445,
      "loss": 5.4635,
      "mean_token_accuracy": 0.17263396084308624,
      "num_tokens": 8219137.0,
      "step": 4460
    },
    {
      "entropy": 5.583094501495362,
      "epoch": 0.3729535583027063,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004990553821645346,
      "loss": 5.4183,
      "mean_token_accuracy": 0.17537619173526764,
      "num_tokens": 8228076.0,
      "step": 4465
    },
    {
      "entropy": 5.435506439208984,
      "epoch": 0.3733711994654193,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000499052655156265,
      "loss": 5.3776,
      "mean_token_accuracy": 0.17335729748010636,
      "num_tokens": 8237581.0,
      "step": 4470
    },
    {
      "entropy": 5.482702732086182,
      "epoch": 0.37378884062813234,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004990499242256839,
      "loss": 5.4332,
      "mean_token_accuracy": 0.16622574329376222,
      "num_tokens": 8247384.0,
      "step": 4475
    },
    {
      "entropy": 5.519318246841431,
      "epoch": 0.3742064817908453,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004990471893728394,
      "loss": 5.4998,
      "mean_token_accuracy": 0.1719047874212265,
      "num_tokens": 8256383.0,
      "step": 4480
    },
    {
      "entropy": 5.478116226196289,
      "epoch": 0.3746241229535583,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000499044450597779,
      "loss": 5.3417,
      "mean_token_accuracy": 0.1772901251912117,
      "num_tokens": 8264848.0,
      "step": 4485
    },
    {
      "entropy": 5.41370005607605,
      "epoch": 0.3750417641162713,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004990417079005508,
      "loss": 5.3956,
      "mean_token_accuracy": 0.16926388293504716,
      "num_tokens": 8274403.0,
      "step": 4490
    },
    {
      "entropy": 5.41335620880127,
      "epoch": 0.3754594052789843,
      "grad_norm": 1.15625,
      "learning_rate": 0.000499038961281203,
      "loss": 5.3134,
      "mean_token_accuracy": 0.17994887828826905,
      "num_tokens": 8282700.0,
      "step": 4495
    },
    {
      "entropy": 5.394307374954224,
      "epoch": 0.3758770464416973,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004990362107397834,
      "loss": 5.3429,
      "mean_token_accuracy": 0.18360423445701599,
      "num_tokens": 8291718.0,
      "step": 4500
    },
    {
      "entropy": 5.424862241744995,
      "epoch": 0.3762946876044103,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004990334562763401,
      "loss": 5.4079,
      "mean_token_accuracy": 0.17094081193208693,
      "num_tokens": 8301113.0,
      "step": 4505
    },
    {
      "entropy": 5.578273153305053,
      "epoch": 0.3767123287671233,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004990306978909216,
      "loss": 5.5072,
      "mean_token_accuracy": 0.16932496577501296,
      "num_tokens": 8310357.0,
      "step": 4510
    },
    {
      "entropy": 5.541999769210816,
      "epoch": 0.3771299699298363,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004990279355835761,
      "loss": 5.4356,
      "mean_token_accuracy": 0.1734210342168808,
      "num_tokens": 8319680.0,
      "step": 4515
    },
    {
      "entropy": 5.518919038772583,
      "epoch": 0.3775476110925493,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004990251693543519,
      "loss": 5.4503,
      "mean_token_accuracy": 0.16900236755609513,
      "num_tokens": 8329127.0,
      "step": 4520
    },
    {
      "entropy": 5.372347497940064,
      "epoch": 0.3779652522552623,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004990223992032975,
      "loss": 5.2786,
      "mean_token_accuracy": 0.1781455770134926,
      "num_tokens": 8338308.0,
      "step": 4525
    },
    {
      "entropy": 5.426207828521728,
      "epoch": 0.37838289341797526,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004990196251304612,
      "loss": 5.3488,
      "mean_token_accuracy": 0.17504175305366515,
      "num_tokens": 8347491.0,
      "step": 4530
    },
    {
      "entropy": 5.359329080581665,
      "epoch": 0.3788005345806883,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000499016847135892,
      "loss": 5.3347,
      "mean_token_accuracy": 0.17484522759914398,
      "num_tokens": 8357112.0,
      "step": 4535
    },
    {
      "entropy": 5.412172460556031,
      "epoch": 0.3792181757434013,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000499014065219638,
      "loss": 5.3399,
      "mean_token_accuracy": 0.17453625351190566,
      "num_tokens": 8366285.0,
      "step": 4540
    },
    {
      "entropy": 5.40165958404541,
      "epoch": 0.37963581690611425,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004990112793817481,
      "loss": 5.2468,
      "mean_token_accuracy": 0.18874731212854384,
      "num_tokens": 8374362.0,
      "step": 4545
    },
    {
      "entropy": 5.430682134628296,
      "epoch": 0.3800534580688273,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004990084896222713,
      "loss": 5.4874,
      "mean_token_accuracy": 0.16845017224550246,
      "num_tokens": 8383297.0,
      "step": 4550
    },
    {
      "entropy": 5.473242664337159,
      "epoch": 0.38047109923154027,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004990056959412562,
      "loss": 5.3569,
      "mean_token_accuracy": 0.18510000109672547,
      "num_tokens": 8391926.0,
      "step": 4555
    },
    {
      "entropy": 5.49044280052185,
      "epoch": 0.38088874039425324,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004990028983387517,
      "loss": 5.3477,
      "mean_token_accuracy": 0.17314921021461488,
      "num_tokens": 8401226.0,
      "step": 4560
    },
    {
      "entropy": 5.439341115951538,
      "epoch": 0.3813063815569663,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004990000968148068,
      "loss": 5.3758,
      "mean_token_accuracy": 0.1755836471915245,
      "num_tokens": 8409966.0,
      "step": 4565
    },
    {
      "entropy": 5.357498550415039,
      "epoch": 0.38172402271967926,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004989972913694707,
      "loss": 5.2265,
      "mean_token_accuracy": 0.18585699796676636,
      "num_tokens": 8418281.0,
      "step": 4570
    },
    {
      "entropy": 5.300945854187011,
      "epoch": 0.38214166388239224,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004989944820027923,
      "loss": 5.3367,
      "mean_token_accuracy": 0.181486676633358,
      "num_tokens": 8427436.0,
      "step": 4575
    },
    {
      "entropy": 5.474956798553467,
      "epoch": 0.38255930504510527,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004989916687148209,
      "loss": 5.3722,
      "mean_token_accuracy": 0.1815724179148674,
      "num_tokens": 8436066.0,
      "step": 4580
    },
    {
      "entropy": 5.464441156387329,
      "epoch": 0.38297694620781825,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004989888515056057,
      "loss": 5.3875,
      "mean_token_accuracy": 0.17339179962873458,
      "num_tokens": 8445572.0,
      "step": 4585
    },
    {
      "entropy": 5.505393123626709,
      "epoch": 0.3833945873705312,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000498986030375196,
      "loss": 5.4213,
      "mean_token_accuracy": 0.17482264935970307,
      "num_tokens": 8454781.0,
      "step": 4590
    },
    {
      "entropy": 5.459455108642578,
      "epoch": 0.38381222853324426,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004989832053236413,
      "loss": 5.4976,
      "mean_token_accuracy": 0.17263346761465073,
      "num_tokens": 8463239.0,
      "step": 4595
    },
    {
      "entropy": 5.538398218154907,
      "epoch": 0.38422986969595724,
      "grad_norm": 1.21875,
      "learning_rate": 0.000498980376350991,
      "loss": 5.3965,
      "mean_token_accuracy": 0.17584720104932786,
      "num_tokens": 8472664.0,
      "step": 4600
    },
    {
      "entropy": 5.519916439056397,
      "epoch": 0.3846475108586702,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004989775434572945,
      "loss": 5.4656,
      "mean_token_accuracy": 0.16939178854227066,
      "num_tokens": 8483149.0,
      "step": 4605
    },
    {
      "entropy": 5.521700716018676,
      "epoch": 0.38506515202138325,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004989747066426015,
      "loss": 5.3947,
      "mean_token_accuracy": 0.17537314295768738,
      "num_tokens": 8492297.0,
      "step": 4610
    },
    {
      "entropy": 5.456653833389282,
      "epoch": 0.38548279318409623,
      "grad_norm": 1.125,
      "learning_rate": 0.0004989718659069618,
      "loss": 5.4753,
      "mean_token_accuracy": 0.1676836743950844,
      "num_tokens": 8502092.0,
      "step": 4615
    },
    {
      "entropy": 5.502790832519532,
      "epoch": 0.3859004343468092,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000498969021250425,
      "loss": 5.3817,
      "mean_token_accuracy": 0.17625278532505034,
      "num_tokens": 8510816.0,
      "step": 4620
    },
    {
      "entropy": 5.453312110900879,
      "epoch": 0.38631807550952224,
      "grad_norm": 1.125,
      "learning_rate": 0.0004989661726730407,
      "loss": 5.4001,
      "mean_token_accuracy": 0.17780623733997344,
      "num_tokens": 8520055.0,
      "step": 4625
    },
    {
      "entropy": 5.481811618804931,
      "epoch": 0.3867357166722352,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004989633201748591,
      "loss": 5.4122,
      "mean_token_accuracy": 0.1767112836241722,
      "num_tokens": 8528271.0,
      "step": 4630
    },
    {
      "entropy": 5.408877563476563,
      "epoch": 0.3871533578349482,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004989604637559299,
      "loss": 5.3158,
      "mean_token_accuracy": 0.18203164488077164,
      "num_tokens": 8537360.0,
      "step": 4635
    },
    {
      "entropy": 5.310453414916992,
      "epoch": 0.38757099899766123,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004989576034163033,
      "loss": 5.2706,
      "mean_token_accuracy": 0.18779359757900238,
      "num_tokens": 8546574.0,
      "step": 4640
    },
    {
      "entropy": 5.427399921417236,
      "epoch": 0.3879886401603742,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004989547391560293,
      "loss": 5.3793,
      "mean_token_accuracy": 0.17438787817955018,
      "num_tokens": 8555581.0,
      "step": 4645
    },
    {
      "entropy": 5.435323333740234,
      "epoch": 0.3884062813230872,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000498951870975158,
      "loss": 5.4074,
      "mean_token_accuracy": 0.17851884365081788,
      "num_tokens": 8564539.0,
      "step": 4650
    },
    {
      "entropy": 5.462069654464722,
      "epoch": 0.3888239224858002,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004989489988737395,
      "loss": 5.4496,
      "mean_token_accuracy": 0.17168653905391693,
      "num_tokens": 8573804.0,
      "step": 4655
    },
    {
      "entropy": 5.404630899429321,
      "epoch": 0.3892415636485132,
      "grad_norm": 1.125,
      "learning_rate": 0.0004989461228518243,
      "loss": 5.3602,
      "mean_token_accuracy": 0.18074435889720916,
      "num_tokens": 8583262.0,
      "step": 4660
    },
    {
      "entropy": 5.411992454528809,
      "epoch": 0.3896592048112262,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004989432429094627,
      "loss": 5.4417,
      "mean_token_accuracy": 0.17372126281261444,
      "num_tokens": 8592763.0,
      "step": 4665
    },
    {
      "entropy": 5.559574317932129,
      "epoch": 0.3900768459739392,
      "grad_norm": 1.125,
      "learning_rate": 0.0004989403590467049,
      "loss": 5.3195,
      "mean_token_accuracy": 0.17951521873474122,
      "num_tokens": 8601884.0,
      "step": 4670
    },
    {
      "entropy": 5.382081413269043,
      "epoch": 0.3904944871366522,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004989374712636017,
      "loss": 5.4297,
      "mean_token_accuracy": 0.17157796323299407,
      "num_tokens": 8611207.0,
      "step": 4675
    },
    {
      "entropy": 5.317033100128174,
      "epoch": 0.39091212829936517,
      "grad_norm": 1.125,
      "learning_rate": 0.0004989345795602036,
      "loss": 5.1908,
      "mean_token_accuracy": 0.19590218216180802,
      "num_tokens": 8620104.0,
      "step": 4680
    },
    {
      "entropy": 5.452140617370605,
      "epoch": 0.3913297694620782,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004989316839365609,
      "loss": 5.4019,
      "mean_token_accuracy": 0.17041265070438386,
      "num_tokens": 8629552.0,
      "step": 4685
    },
    {
      "entropy": 5.435151386260986,
      "epoch": 0.3917474106247912,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004989287843927246,
      "loss": 5.3305,
      "mean_token_accuracy": 0.17590319067239762,
      "num_tokens": 8638961.0,
      "step": 4690
    },
    {
      "entropy": 5.419300651550293,
      "epoch": 0.39216505178750416,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004989258809287455,
      "loss": 5.3655,
      "mean_token_accuracy": 0.17324947267770768,
      "num_tokens": 8648051.0,
      "step": 4695
    },
    {
      "entropy": 5.4347825050354,
      "epoch": 0.3925826929502172,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004989229735446743,
      "loss": 5.3435,
      "mean_token_accuracy": 0.17478834241628646,
      "num_tokens": 8656863.0,
      "step": 4700
    },
    {
      "entropy": 5.251941823959351,
      "epoch": 0.39300033411293017,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004989200622405619,
      "loss": 5.1806,
      "mean_token_accuracy": 0.18895964771509172,
      "num_tokens": 8665455.0,
      "step": 4705
    },
    {
      "entropy": 5.324203968048096,
      "epoch": 0.39341797527564315,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004989171470164592,
      "loss": 5.2925,
      "mean_token_accuracy": 0.18248264938592912,
      "num_tokens": 8674119.0,
      "step": 4710
    },
    {
      "entropy": 5.469828939437866,
      "epoch": 0.3938356164383562,
      "grad_norm": 1.125,
      "learning_rate": 0.0004989142278724173,
      "loss": 5.3925,
      "mean_token_accuracy": 0.1724318742752075,
      "num_tokens": 8682979.0,
      "step": 4715
    },
    {
      "entropy": 5.469657611846924,
      "epoch": 0.39425325760106916,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004989113048084874,
      "loss": 5.2976,
      "mean_token_accuracy": 0.1702787160873413,
      "num_tokens": 8692392.0,
      "step": 4720
    },
    {
      "entropy": 5.318474864959716,
      "epoch": 0.39467089876378214,
      "grad_norm": 1.125,
      "learning_rate": 0.0004989083778247206,
      "loss": 5.358,
      "mean_token_accuracy": 0.17889411002397537,
      "num_tokens": 8700932.0,
      "step": 4725
    },
    {
      "entropy": 5.550222682952881,
      "epoch": 0.3950885399264952,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000498905446921168,
      "loss": 5.5122,
      "mean_token_accuracy": 0.1707063190639019,
      "num_tokens": 8710683.0,
      "step": 4730
    },
    {
      "entropy": 5.381403684616089,
      "epoch": 0.39550618108920815,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004989025120978812,
      "loss": 5.3468,
      "mean_token_accuracy": 0.17637370228767396,
      "num_tokens": 8719939.0,
      "step": 4735
    },
    {
      "entropy": 5.469001817703247,
      "epoch": 0.39592382225192113,
      "grad_norm": 1.125,
      "learning_rate": 0.0004988995733549113,
      "loss": 5.4081,
      "mean_token_accuracy": 0.17335674911737442,
      "num_tokens": 8729275.0,
      "step": 4740
    },
    {
      "entropy": 5.355962038040161,
      "epoch": 0.39634146341463417,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004988966306923099,
      "loss": 5.2687,
      "mean_token_accuracy": 0.180453297495842,
      "num_tokens": 8738048.0,
      "step": 4745
    },
    {
      "entropy": 5.426445722579956,
      "epoch": 0.39675910457734714,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004988936841101284,
      "loss": 5.3557,
      "mean_token_accuracy": 0.17539500296115876,
      "num_tokens": 8746967.0,
      "step": 4750
    },
    {
      "entropy": 5.358024406433105,
      "epoch": 0.3971767457400601,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004988907336084186,
      "loss": 5.2957,
      "mean_token_accuracy": 0.18092308789491654,
      "num_tokens": 8754908.0,
      "step": 4755
    },
    {
      "entropy": 5.396121072769165,
      "epoch": 0.39759438690277316,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004988877791872319,
      "loss": 5.3433,
      "mean_token_accuracy": 0.18545244336128236,
      "num_tokens": 8764489.0,
      "step": 4760
    },
    {
      "entropy": 5.565548896789551,
      "epoch": 0.39801202806548613,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004988848208466202,
      "loss": 5.5031,
      "mean_token_accuracy": 0.15702270716428757,
      "num_tokens": 8774365.0,
      "step": 4765
    },
    {
      "entropy": 5.449076223373413,
      "epoch": 0.3984296692281991,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004988818585866351,
      "loss": 5.3008,
      "mean_token_accuracy": 0.1790413498878479,
      "num_tokens": 8783538.0,
      "step": 4770
    },
    {
      "entropy": 5.3570427894592285,
      "epoch": 0.39884731039091215,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004988788924073288,
      "loss": 5.4084,
      "mean_token_accuracy": 0.17840720862150192,
      "num_tokens": 8792171.0,
      "step": 4775
    },
    {
      "entropy": 5.432498550415039,
      "epoch": 0.3992649515536251,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004988759223087528,
      "loss": 5.3993,
      "mean_token_accuracy": 0.1746385797858238,
      "num_tokens": 8801325.0,
      "step": 4780
    },
    {
      "entropy": 5.42991533279419,
      "epoch": 0.3996825927163381,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004988729482909595,
      "loss": 5.3439,
      "mean_token_accuracy": 0.1762702226638794,
      "num_tokens": 8810566.0,
      "step": 4785
    },
    {
      "entropy": 5.296753215789795,
      "epoch": 0.40010023387905114,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004988699703540006,
      "loss": 5.2812,
      "mean_token_accuracy": 0.1823352664709091,
      "num_tokens": 8819735.0,
      "step": 4790
    },
    {
      "entropy": 5.402335977554321,
      "epoch": 0.4005178750417641,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004988669884979284,
      "loss": 5.4222,
      "mean_token_accuracy": 0.1739557906985283,
      "num_tokens": 8828820.0,
      "step": 4795
    },
    {
      "entropy": 5.5007377624511715,
      "epoch": 0.4009355162044771,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004988640027227951,
      "loss": 5.3161,
      "mean_token_accuracy": 0.17225112169981002,
      "num_tokens": 8837858.0,
      "step": 4800
    },
    {
      "entropy": 5.402754068374634,
      "epoch": 0.4013531573671901,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000498861013028653,
      "loss": 5.274,
      "mean_token_accuracy": 0.18401779979467392,
      "num_tokens": 8847172.0,
      "step": 4805
    },
    {
      "entropy": 5.359449672698974,
      "epoch": 0.4017707985299031,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004988580194155544,
      "loss": 5.323,
      "mean_token_accuracy": 0.17467405050992965,
      "num_tokens": 8856700.0,
      "step": 4810
    },
    {
      "entropy": 5.297247648239136,
      "epoch": 0.4021884396926161,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004988550218835517,
      "loss": 5.2096,
      "mean_token_accuracy": 0.18494719564914702,
      "num_tokens": 8866108.0,
      "step": 4815
    },
    {
      "entropy": 5.347088050842285,
      "epoch": 0.4026060808553291,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004988520204326975,
      "loss": 5.3029,
      "mean_token_accuracy": 0.18589287698268891,
      "num_tokens": 8874833.0,
      "step": 4820
    },
    {
      "entropy": 5.445062351226807,
      "epoch": 0.4030237220180421,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004988490150630441,
      "loss": 5.3936,
      "mean_token_accuracy": 0.17721084058284758,
      "num_tokens": 8884211.0,
      "step": 4825
    },
    {
      "entropy": 5.500857543945313,
      "epoch": 0.4034413631807551,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004988460057746443,
      "loss": 5.4927,
      "mean_token_accuracy": 0.16921537965536118,
      "num_tokens": 8893443.0,
      "step": 4830
    },
    {
      "entropy": 5.369778633117676,
      "epoch": 0.4038590043434681,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004988429925675507,
      "loss": 5.3515,
      "mean_token_accuracy": 0.17862945795059204,
      "num_tokens": 8903544.0,
      "step": 4835
    },
    {
      "entropy": 5.46190824508667,
      "epoch": 0.4042766455061811,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004988399754418161,
      "loss": 5.366,
      "mean_token_accuracy": 0.1793976604938507,
      "num_tokens": 8912270.0,
      "step": 4840
    },
    {
      "entropy": 5.537768125534058,
      "epoch": 0.40469428666889407,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004988369543974932,
      "loss": 5.461,
      "mean_token_accuracy": 0.1689763829112053,
      "num_tokens": 8922155.0,
      "step": 4845
    },
    {
      "entropy": 5.473590850830078,
      "epoch": 0.4051119278316071,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004988339294346351,
      "loss": 5.4562,
      "mean_token_accuracy": 0.17304005473852158,
      "num_tokens": 8931844.0,
      "step": 4850
    },
    {
      "entropy": 5.4567924499511715,
      "epoch": 0.4055295689943201,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004988309005532946,
      "loss": 5.3169,
      "mean_token_accuracy": 0.18166026920080186,
      "num_tokens": 8940874.0,
      "step": 4855
    },
    {
      "entropy": 5.412811899185181,
      "epoch": 0.40594721015703306,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004988278677535247,
      "loss": 5.3881,
      "mean_token_accuracy": 0.18285960853099822,
      "num_tokens": 8950108.0,
      "step": 4860
    },
    {
      "entropy": 5.473009872436523,
      "epoch": 0.4063648513197461,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004988248310353788,
      "loss": 5.4083,
      "mean_token_accuracy": 0.17212433815002443,
      "num_tokens": 8960610.0,
      "step": 4865
    },
    {
      "entropy": 5.355665683746338,
      "epoch": 0.40678249248245907,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004988217903989095,
      "loss": 5.325,
      "mean_token_accuracy": 0.17474715560674667,
      "num_tokens": 8969853.0,
      "step": 4870
    },
    {
      "entropy": 5.465075540542602,
      "epoch": 0.40720013364517205,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004988187458441704,
      "loss": 5.3095,
      "mean_token_accuracy": 0.17334848046302795,
      "num_tokens": 8979800.0,
      "step": 4875
    },
    {
      "entropy": 5.447531795501709,
      "epoch": 0.4076177748078851,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004988156973712149,
      "loss": 5.3784,
      "mean_token_accuracy": 0.1711223915219307,
      "num_tokens": 8988667.0,
      "step": 4880
    },
    {
      "entropy": 5.289280891418457,
      "epoch": 0.40803541597059806,
      "grad_norm": 1.25,
      "learning_rate": 0.0004988126449800961,
      "loss": 5.2575,
      "mean_token_accuracy": 0.1802578404545784,
      "num_tokens": 8997385.0,
      "step": 4885
    },
    {
      "entropy": 5.353163862228394,
      "epoch": 0.40845305713331104,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004988095886708676,
      "loss": 5.2729,
      "mean_token_accuracy": 0.18317050039768218,
      "num_tokens": 9005999.0,
      "step": 4890
    },
    {
      "entropy": 5.447890090942383,
      "epoch": 0.40887069829602407,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004988065284435828,
      "loss": 5.4104,
      "mean_token_accuracy": 0.18185970485210418,
      "num_tokens": 9014837.0,
      "step": 4895
    },
    {
      "entropy": 5.505316591262817,
      "epoch": 0.40928833945873705,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004988034642982954,
      "loss": 5.4413,
      "mean_token_accuracy": 0.1804886966943741,
      "num_tokens": 9023956.0,
      "step": 4900
    },
    {
      "entropy": 5.510533952713013,
      "epoch": 0.40970598062145,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004988003962350589,
      "loss": 5.4033,
      "mean_token_accuracy": 0.16933871060609818,
      "num_tokens": 9032945.0,
      "step": 4905
    },
    {
      "entropy": 5.41035795211792,
      "epoch": 0.41012362178416306,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000498797324253927,
      "loss": 5.3349,
      "mean_token_accuracy": 0.1821726843714714,
      "num_tokens": 9042109.0,
      "step": 4910
    },
    {
      "entropy": 5.327030658721924,
      "epoch": 0.41054126294687604,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004987942483549537,
      "loss": 5.3237,
      "mean_token_accuracy": 0.17505295425653458,
      "num_tokens": 9052210.0,
      "step": 4915
    },
    {
      "entropy": 5.460620880126953,
      "epoch": 0.410958904109589,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004987911685381926,
      "loss": 5.2838,
      "mean_token_accuracy": 0.18444007337093354,
      "num_tokens": 9060089.0,
      "step": 4920
    },
    {
      "entropy": 5.486596584320068,
      "epoch": 0.41137654527230205,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004987880848036977,
      "loss": 5.3995,
      "mean_token_accuracy": 0.17101875692605972,
      "num_tokens": 9069235.0,
      "step": 4925
    },
    {
      "entropy": 5.3632927417755125,
      "epoch": 0.41179418643501503,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004987849971515232,
      "loss": 5.3398,
      "mean_token_accuracy": 0.18175618052482606,
      "num_tokens": 9077509.0,
      "step": 4930
    },
    {
      "entropy": 5.4160401821136475,
      "epoch": 0.412211827597728,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004987819055817227,
      "loss": 5.2523,
      "mean_token_accuracy": 0.18890025466680527,
      "num_tokens": 9085493.0,
      "step": 4935
    },
    {
      "entropy": 5.265770339965821,
      "epoch": 0.41262946876044104,
      "grad_norm": 1.125,
      "learning_rate": 0.0004987788100943507,
      "loss": 5.2334,
      "mean_token_accuracy": 0.18305041491985322,
      "num_tokens": 9094748.0,
      "step": 4940
    },
    {
      "entropy": 5.399051094055176,
      "epoch": 0.413047109923154,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004987757106894613,
      "loss": 5.2969,
      "mean_token_accuracy": 0.18176668584346772,
      "num_tokens": 9103678.0,
      "step": 4945
    },
    {
      "entropy": 5.412854290008545,
      "epoch": 0.413464751085867,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004987726073671086,
      "loss": 5.3701,
      "mean_token_accuracy": 0.1806068480014801,
      "num_tokens": 9112851.0,
      "step": 4950
    },
    {
      "entropy": 5.53558030128479,
      "epoch": 0.41388239224858003,
      "grad_norm": 1.125,
      "learning_rate": 0.0004987695001273471,
      "loss": 5.5739,
      "mean_token_accuracy": 0.16550440490245819,
      "num_tokens": 9122106.0,
      "step": 4955
    },
    {
      "entropy": 5.442167043685913,
      "epoch": 0.414300033411293,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004987663889702313,
      "loss": 5.2991,
      "mean_token_accuracy": 0.18062672168016433,
      "num_tokens": 9131580.0,
      "step": 4960
    },
    {
      "entropy": 5.481163120269775,
      "epoch": 0.414717674574006,
      "grad_norm": 1.125,
      "learning_rate": 0.0004987632738958155,
      "loss": 5.4317,
      "mean_token_accuracy": 0.17543174624443053,
      "num_tokens": 9141642.0,
      "step": 4965
    },
    {
      "entropy": 5.553693437576294,
      "epoch": 0.415135315736719,
      "grad_norm": 1.125,
      "learning_rate": 0.0004987601549041541,
      "loss": 5.482,
      "mean_token_accuracy": 0.16618218123912812,
      "num_tokens": 9151192.0,
      "step": 4970
    },
    {
      "entropy": 5.424982213973999,
      "epoch": 0.415552956899432,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004987570319953018,
      "loss": 5.3664,
      "mean_token_accuracy": 0.17582743167877196,
      "num_tokens": 9161316.0,
      "step": 4975
    },
    {
      "entropy": 5.487713193893432,
      "epoch": 0.415970598062145,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004987539051693134,
      "loss": 5.3789,
      "mean_token_accuracy": 0.170578333735466,
      "num_tokens": 9169535.0,
      "step": 4980
    },
    {
      "entropy": 5.433673286437989,
      "epoch": 0.416388239224858,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004987507744262436,
      "loss": 5.401,
      "mean_token_accuracy": 0.1698138326406479,
      "num_tokens": 9178187.0,
      "step": 4985
    },
    {
      "entropy": 5.338801288604737,
      "epoch": 0.416805880387571,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004987476397661472,
      "loss": 5.2987,
      "mean_token_accuracy": 0.17456402033567428,
      "num_tokens": 9186342.0,
      "step": 4990
    },
    {
      "entropy": 5.470065784454346,
      "epoch": 0.41722352155028397,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004987445011890789,
      "loss": 5.4232,
      "mean_token_accuracy": 0.17456976175308228,
      "num_tokens": 9196470.0,
      "step": 4995
    },
    {
      "entropy": 5.415784645080566,
      "epoch": 0.417641162712997,
      "grad_norm": 1.125,
      "learning_rate": 0.000498741358695094,
      "loss": 5.3299,
      "mean_token_accuracy": 0.17888012081384658,
      "num_tokens": 9206587.0,
      "step": 5000
    },
    {
      "entropy": 5.377053928375244,
      "epoch": 0.41805880387571,
      "grad_norm": 1.078125,
      "learning_rate": 0.000498738212284247,
      "loss": 5.3267,
      "mean_token_accuracy": 0.17406473457813262,
      "num_tokens": 9216459.0,
      "step": 5005
    },
    {
      "entropy": 5.556557559967041,
      "epoch": 0.41847644503842296,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004987350619565934,
      "loss": 5.44,
      "mean_token_accuracy": 0.17223341315984725,
      "num_tokens": 9227301.0,
      "step": 5010
    },
    {
      "entropy": 5.4989378452301025,
      "epoch": 0.418894086201136,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004987319077121884,
      "loss": 5.3702,
      "mean_token_accuracy": 0.16402043551206588,
      "num_tokens": 9236577.0,
      "step": 5015
    },
    {
      "entropy": 5.350860643386841,
      "epoch": 0.419311727363849,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004987287495510869,
      "loss": 5.362,
      "mean_token_accuracy": 0.17799901366233825,
      "num_tokens": 9246451.0,
      "step": 5020
    },
    {
      "entropy": 5.3903028011322025,
      "epoch": 0.41972936852656195,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004987255874733444,
      "loss": 5.244,
      "mean_token_accuracy": 0.1740152657032013,
      "num_tokens": 9256108.0,
      "step": 5025
    },
    {
      "entropy": 5.315888261795044,
      "epoch": 0.420147009689275,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004987224214790161,
      "loss": 5.3004,
      "mean_token_accuracy": 0.1798888012766838,
      "num_tokens": 9265173.0,
      "step": 5030
    },
    {
      "entropy": 5.426610898971558,
      "epoch": 0.42056465085198796,
      "grad_norm": 1.125,
      "learning_rate": 0.0004987192515681575,
      "loss": 5.4024,
      "mean_token_accuracy": 0.17473443150520324,
      "num_tokens": 9273979.0,
      "step": 5035
    },
    {
      "entropy": 5.392903518676758,
      "epoch": 0.42098229201470094,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004987160777408243,
      "loss": 5.3462,
      "mean_token_accuracy": 0.17106169909238816,
      "num_tokens": 9281893.0,
      "step": 5040
    },
    {
      "entropy": 5.513010692596436,
      "epoch": 0.421399933177414,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004987128999970717,
      "loss": 5.4618,
      "mean_token_accuracy": 0.1757590688765049,
      "num_tokens": 9290799.0,
      "step": 5045
    },
    {
      "entropy": 5.558505439758301,
      "epoch": 0.42181757434012696,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004987097183369555,
      "loss": 5.4302,
      "mean_token_accuracy": 0.16776997298002244,
      "num_tokens": 9300562.0,
      "step": 5050
    },
    {
      "entropy": 5.30872278213501,
      "epoch": 0.42223521550283993,
      "grad_norm": 1.125,
      "learning_rate": 0.0004987065327605315,
      "loss": 5.2133,
      "mean_token_accuracy": 0.18027304857969284,
      "num_tokens": 9308889.0,
      "step": 5055
    },
    {
      "entropy": 5.29863543510437,
      "epoch": 0.42265285666555297,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004987033432678553,
      "loss": 5.2958,
      "mean_token_accuracy": 0.18111589998006822,
      "num_tokens": 9317957.0,
      "step": 5060
    },
    {
      "entropy": 5.51900086402893,
      "epoch": 0.42307049782826595,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004987001498589828,
      "loss": 5.4949,
      "mean_token_accuracy": 0.16496727764606475,
      "num_tokens": 9327849.0,
      "step": 5065
    },
    {
      "entropy": 5.506615066528321,
      "epoch": 0.4234881389909789,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00049869695253397,
      "loss": 5.3484,
      "mean_token_accuracy": 0.1687320351600647,
      "num_tokens": 9337165.0,
      "step": 5070
    },
    {
      "entropy": 5.390972232818603,
      "epoch": 0.42390578015369196,
      "grad_norm": 1.125,
      "learning_rate": 0.0004986937512928727,
      "loss": 5.2594,
      "mean_token_accuracy": 0.18033367842435838,
      "num_tokens": 9346817.0,
      "step": 5075
    },
    {
      "entropy": 5.45494441986084,
      "epoch": 0.42432342131640494,
      "grad_norm": 1.125,
      "learning_rate": 0.000498690546135747,
      "loss": 5.3816,
      "mean_token_accuracy": 0.17511844784021377,
      "num_tokens": 9356420.0,
      "step": 5080
    },
    {
      "entropy": 5.33007960319519,
      "epoch": 0.4247410624791179,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004986873370626491,
      "loss": 5.3068,
      "mean_token_accuracy": 0.18187834471464157,
      "num_tokens": 9365580.0,
      "step": 5085
    },
    {
      "entropy": 5.42194504737854,
      "epoch": 0.42515870364183095,
      "grad_norm": 1.15625,
      "learning_rate": 0.000498684124073635,
      "loss": 5.3709,
      "mean_token_accuracy": 0.16973898261785508,
      "num_tokens": 9375116.0,
      "step": 5090
    },
    {
      "entropy": 5.4432093620300295,
      "epoch": 0.4255763448045439,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004986809071687612,
      "loss": 5.336,
      "mean_token_accuracy": 0.18046289086341857,
      "num_tokens": 9384129.0,
      "step": 5095
    },
    {
      "entropy": 5.370714426040649,
      "epoch": 0.4259939859672569,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004986776863480838,
      "loss": 5.2769,
      "mean_token_accuracy": 0.17603334188461303,
      "num_tokens": 9392757.0,
      "step": 5100
    },
    {
      "entropy": 5.287596130371094,
      "epoch": 0.42641162712996994,
      "grad_norm": 1.125,
      "learning_rate": 0.0004986744616116593,
      "loss": 5.2615,
      "mean_token_accuracy": 0.18092862516641617,
      "num_tokens": 9401753.0,
      "step": 5105
    },
    {
      "entropy": 5.397714376449585,
      "epoch": 0.4268292682926829,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000498671232959544,
      "loss": 5.4721,
      "mean_token_accuracy": 0.166457898914814,
      "num_tokens": 9411400.0,
      "step": 5110
    },
    {
      "entropy": 5.519891691207886,
      "epoch": 0.42724690945539595,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004986680003917947,
      "loss": 5.3158,
      "mean_token_accuracy": 0.1766785129904747,
      "num_tokens": 9420206.0,
      "step": 5115
    },
    {
      "entropy": 5.367687368392945,
      "epoch": 0.42766455061810893,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004986647639084677,
      "loss": 5.3018,
      "mean_token_accuracy": 0.17577457129955293,
      "num_tokens": 9428418.0,
      "step": 5120
    },
    {
      "entropy": 5.389084577560425,
      "epoch": 0.4280821917808219,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00049866152350962,
      "loss": 5.3707,
      "mean_token_accuracy": 0.17752706706523896,
      "num_tokens": 9437624.0,
      "step": 5125
    },
    {
      "entropy": 5.434055614471435,
      "epoch": 0.42849983294353494,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004986582791953078,
      "loss": 5.2778,
      "mean_token_accuracy": 0.18143245875835418,
      "num_tokens": 9446857.0,
      "step": 5130
    },
    {
      "entropy": 5.367829656600952,
      "epoch": 0.4289174741062479,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004986550309655884,
      "loss": 5.3738,
      "mean_token_accuracy": 0.17304729223251342,
      "num_tokens": 9456024.0,
      "step": 5135
    },
    {
      "entropy": 5.361046934127808,
      "epoch": 0.4293351152689609,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004986517788205184,
      "loss": 5.2657,
      "mean_token_accuracy": 0.18356824517250062,
      "num_tokens": 9464018.0,
      "step": 5140
    },
    {
      "entropy": 5.348368215560913,
      "epoch": 0.42975275643167393,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004986485227601548,
      "loss": 5.3886,
      "mean_token_accuracy": 0.17977178692817689,
      "num_tokens": 9472430.0,
      "step": 5145
    },
    {
      "entropy": 5.387197065353393,
      "epoch": 0.4301703975943869,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004986452627845547,
      "loss": 5.3092,
      "mean_token_accuracy": 0.182216514647007,
      "num_tokens": 9480592.0,
      "step": 5150
    },
    {
      "entropy": 5.368278741836548,
      "epoch": 0.4305880387570999,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000498641998893775,
      "loss": 5.2648,
      "mean_token_accuracy": 0.1800767406821251,
      "num_tokens": 9489162.0,
      "step": 5155
    },
    {
      "entropy": 5.408619976043701,
      "epoch": 0.4310056799198129,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004986387310878728,
      "loss": 5.3317,
      "mean_token_accuracy": 0.17703358680009842,
      "num_tokens": 9499210.0,
      "step": 5160
    },
    {
      "entropy": 5.426770877838135,
      "epoch": 0.4314233210825259,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004986354593669056,
      "loss": 5.3683,
      "mean_token_accuracy": 0.18122825920581817,
      "num_tokens": 9508062.0,
      "step": 5165
    },
    {
      "entropy": 5.43633918762207,
      "epoch": 0.4318409622452389,
      "grad_norm": 1.125,
      "learning_rate": 0.0004986321837309304,
      "loss": 5.3037,
      "mean_token_accuracy": 0.18372174352407455,
      "num_tokens": 9516756.0,
      "step": 5170
    },
    {
      "entropy": 5.371243238449097,
      "epoch": 0.4322586034079519,
      "grad_norm": 1.125,
      "learning_rate": 0.0004986289041800046,
      "loss": 5.3046,
      "mean_token_accuracy": 0.1816995456814766,
      "num_tokens": 9525305.0,
      "step": 5175
    },
    {
      "entropy": 5.390642595291138,
      "epoch": 0.4326762445706649,
      "grad_norm": 1.375,
      "learning_rate": 0.0004986256207141857,
      "loss": 5.2849,
      "mean_token_accuracy": 0.18394368886947632,
      "num_tokens": 9534310.0,
      "step": 5180
    },
    {
      "entropy": 5.5049614906311035,
      "epoch": 0.43309388573337787,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004986223333335311,
      "loss": 5.3968,
      "mean_token_accuracy": 0.17676397114992143,
      "num_tokens": 9543648.0,
      "step": 5185
    },
    {
      "entropy": 5.33973445892334,
      "epoch": 0.4335115268960909,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004986190420380985,
      "loss": 5.3115,
      "mean_token_accuracy": 0.1795612543821335,
      "num_tokens": 9553081.0,
      "step": 5190
    },
    {
      "entropy": 5.451060962677002,
      "epoch": 0.4339291680588039,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004986157468279453,
      "loss": 5.3564,
      "mean_token_accuracy": 0.17509184926748275,
      "num_tokens": 9563184.0,
      "step": 5195
    },
    {
      "entropy": 5.426450729370117,
      "epoch": 0.43434680922151686,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004986124477031293,
      "loss": 5.3037,
      "mean_token_accuracy": 0.17883535772562026,
      "num_tokens": 9572550.0,
      "step": 5200
    },
    {
      "entropy": 5.3502343654632565,
      "epoch": 0.4347644503842299,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004986091446637083,
      "loss": 5.357,
      "mean_token_accuracy": 0.17619699239730835,
      "num_tokens": 9580900.0,
      "step": 5205
    },
    {
      "entropy": 5.457106208801269,
      "epoch": 0.4351820915469429,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00049860583770974,
      "loss": 5.3515,
      "mean_token_accuracy": 0.18001936376094818,
      "num_tokens": 9589340.0,
      "step": 5210
    },
    {
      "entropy": 5.472261667251587,
      "epoch": 0.43559973270965585,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004986025268412823,
      "loss": 5.471,
      "mean_token_accuracy": 0.17389730960130692,
      "num_tokens": 9599316.0,
      "step": 5215
    },
    {
      "entropy": 5.452399969100952,
      "epoch": 0.4360173738723689,
      "grad_norm": 1.125,
      "learning_rate": 0.0004985992120583933,
      "loss": 5.3525,
      "mean_token_accuracy": 0.17356273531913757,
      "num_tokens": 9609350.0,
      "step": 5220
    },
    {
      "entropy": 5.474084186553955,
      "epoch": 0.43643501503508186,
      "grad_norm": 1.15625,
      "learning_rate": 0.000498595893361131,
      "loss": 5.3968,
      "mean_token_accuracy": 0.18037155270576477,
      "num_tokens": 9618491.0,
      "step": 5225
    },
    {
      "entropy": 5.49598331451416,
      "epoch": 0.43685265619779484,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004985925707495533,
      "loss": 5.4735,
      "mean_token_accuracy": 0.1682824060320854,
      "num_tokens": 9629613.0,
      "step": 5230
    },
    {
      "entropy": 5.384536600112915,
      "epoch": 0.4372702973605079,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004985892442237186,
      "loss": 5.3269,
      "mean_token_accuracy": 0.1817545160651207,
      "num_tokens": 9638035.0,
      "step": 5235
    },
    {
      "entropy": 5.437534236907959,
      "epoch": 0.43768793852322085,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004985859137836851,
      "loss": 5.4153,
      "mean_token_accuracy": 0.16965906023979188,
      "num_tokens": 9647378.0,
      "step": 5240
    },
    {
      "entropy": 5.478408432006836,
      "epoch": 0.43810557968593383,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004985825794295109,
      "loss": 5.3166,
      "mean_token_accuracy": 0.1800900012254715,
      "num_tokens": 9655728.0,
      "step": 5245
    },
    {
      "entropy": 5.413166236877442,
      "epoch": 0.43852322084864687,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004985792411612546,
      "loss": 5.3675,
      "mean_token_accuracy": 0.18087270259857177,
      "num_tokens": 9665063.0,
      "step": 5250
    },
    {
      "entropy": 5.447279405593872,
      "epoch": 0.43894086201135984,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004985758989789746,
      "loss": 5.3492,
      "mean_token_accuracy": 0.17699085175991058,
      "num_tokens": 9674406.0,
      "step": 5255
    },
    {
      "entropy": 5.343826150894165,
      "epoch": 0.4393585031740728,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004985725528827293,
      "loss": 5.3113,
      "mean_token_accuracy": 0.1798545628786087,
      "num_tokens": 9683303.0,
      "step": 5260
    },
    {
      "entropy": 5.417714405059814,
      "epoch": 0.43977614433678586,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004985692028725774,
      "loss": 5.3111,
      "mean_token_accuracy": 0.17840295135974885,
      "num_tokens": 9692938.0,
      "step": 5265
    },
    {
      "entropy": 5.446576070785523,
      "epoch": 0.44019378549949884,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004985658489485774,
      "loss": 5.3851,
      "mean_token_accuracy": 0.176699498295784,
      "num_tokens": 9702192.0,
      "step": 5270
    },
    {
      "entropy": 5.409464597702026,
      "epoch": 0.4406114266622118,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004985624911107882,
      "loss": 5.358,
      "mean_token_accuracy": 0.17101490646600723,
      "num_tokens": 9711301.0,
      "step": 5275
    },
    {
      "entropy": 5.3845923900604244,
      "epoch": 0.44102906782492485,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004985591293592685,
      "loss": 5.3544,
      "mean_token_accuracy": 0.17416053116321564,
      "num_tokens": 9720768.0,
      "step": 5280
    },
    {
      "entropy": 5.47494215965271,
      "epoch": 0.4414467089876378,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004985557636940771,
      "loss": 5.3737,
      "mean_token_accuracy": 0.17790818065404893,
      "num_tokens": 9730007.0,
      "step": 5285
    },
    {
      "entropy": 5.472876501083374,
      "epoch": 0.4418643501503508,
      "grad_norm": 1.234375,
      "learning_rate": 0.000498552394115273,
      "loss": 5.4244,
      "mean_token_accuracy": 0.17267797291278839,
      "num_tokens": 9739546.0,
      "step": 5290
    },
    {
      "entropy": 5.391746282577515,
      "epoch": 0.44228199131306384,
      "grad_norm": 1.25,
      "learning_rate": 0.0004985490206229151,
      "loss": 5.3165,
      "mean_token_accuracy": 0.18015623986721038,
      "num_tokens": 9749615.0,
      "step": 5295
    },
    {
      "entropy": 5.46711368560791,
      "epoch": 0.4426996324757768,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004985456432170626,
      "loss": 5.4128,
      "mean_token_accuracy": 0.17161605805158614,
      "num_tokens": 9759285.0,
      "step": 5300
    },
    {
      "entropy": 5.372819995880127,
      "epoch": 0.4431172736384898,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004985422618977744,
      "loss": 5.2791,
      "mean_token_accuracy": 0.17291145026683807,
      "num_tokens": 9769396.0,
      "step": 5305
    },
    {
      "entropy": 5.423769426345825,
      "epoch": 0.44353491480120283,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00049853887666511,
      "loss": 5.4033,
      "mean_token_accuracy": 0.17350405901670457,
      "num_tokens": 9779632.0,
      "step": 5310
    },
    {
      "entropy": 5.483930587768555,
      "epoch": 0.4439525559639158,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004985354875191283,
      "loss": 5.3246,
      "mean_token_accuracy": 0.17986151874065398,
      "num_tokens": 9788090.0,
      "step": 5315
    },
    {
      "entropy": 5.325446748733521,
      "epoch": 0.4443701971266288,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004985320944598889,
      "loss": 5.2383,
      "mean_token_accuracy": 0.18873758614063263,
      "num_tokens": 9795844.0,
      "step": 5320
    },
    {
      "entropy": 5.256659317016601,
      "epoch": 0.4447878382893418,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004985286974874512,
      "loss": 5.2251,
      "mean_token_accuracy": 0.18810541033744813,
      "num_tokens": 9805119.0,
      "step": 5325
    },
    {
      "entropy": 5.440543508529663,
      "epoch": 0.4452054794520548,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004985252966018745,
      "loss": 5.3417,
      "mean_token_accuracy": 0.18258800655603408,
      "num_tokens": 9813842.0,
      "step": 5330
    },
    {
      "entropy": 5.519575881958008,
      "epoch": 0.4456231206147678,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004985218918032184,
      "loss": 5.3465,
      "mean_token_accuracy": 0.17161140143871306,
      "num_tokens": 9822872.0,
      "step": 5335
    },
    {
      "entropy": 5.438978767395019,
      "epoch": 0.4460407617774808,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004985184830915426,
      "loss": 5.4609,
      "mean_token_accuracy": 0.1676907256245613,
      "num_tokens": 9832710.0,
      "step": 5340
    },
    {
      "entropy": 5.4792500019073485,
      "epoch": 0.4464584029401938,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004985150704669067,
      "loss": 5.3657,
      "mean_token_accuracy": 0.1726950615644455,
      "num_tokens": 9841047.0,
      "step": 5345
    },
    {
      "entropy": 5.445040225982666,
      "epoch": 0.44687604410290677,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004985116539293704,
      "loss": 5.3323,
      "mean_token_accuracy": 0.1756501317024231,
      "num_tokens": 9850168.0,
      "step": 5350
    },
    {
      "entropy": 5.390517377853394,
      "epoch": 0.4472936852656198,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004985082334789936,
      "loss": 5.3849,
      "mean_token_accuracy": 0.1720145031809807,
      "num_tokens": 9859550.0,
      "step": 5355
    },
    {
      "entropy": 5.530892658233642,
      "epoch": 0.4477113264283328,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004985048091158361,
      "loss": 5.3353,
      "mean_token_accuracy": 0.17469432651996614,
      "num_tokens": 9868317.0,
      "step": 5360
    },
    {
      "entropy": 5.526041316986084,
      "epoch": 0.44812896759104576,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004985013808399578,
      "loss": 5.5569,
      "mean_token_accuracy": 0.15901562422513962,
      "num_tokens": 9878421.0,
      "step": 5365
    },
    {
      "entropy": 5.395598411560059,
      "epoch": 0.4485466087537588,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004984979486514189,
      "loss": 5.2805,
      "mean_token_accuracy": 0.1847289741039276,
      "num_tokens": 9887213.0,
      "step": 5370
    },
    {
      "entropy": 5.380127382278443,
      "epoch": 0.44896424991647177,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004984945125502795,
      "loss": 5.4179,
      "mean_token_accuracy": 0.17433496117591857,
      "num_tokens": 9897274.0,
      "step": 5375
    },
    {
      "entropy": 5.3499444961547855,
      "epoch": 0.44938189107918475,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004984910725365994,
      "loss": 5.3268,
      "mean_token_accuracy": 0.17421112805604935,
      "num_tokens": 9906765.0,
      "step": 5380
    },
    {
      "entropy": 5.537207746505738,
      "epoch": 0.4497995322418978,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004984876286104392,
      "loss": 5.3473,
      "mean_token_accuracy": 0.17053734958171846,
      "num_tokens": 9916149.0,
      "step": 5385
    },
    {
      "entropy": 5.396920490264892,
      "epoch": 0.45021717340461076,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004984841807718591,
      "loss": 5.3371,
      "mean_token_accuracy": 0.17592400908470154,
      "num_tokens": 9925657.0,
      "step": 5390
    },
    {
      "entropy": 5.375749349594116,
      "epoch": 0.45063481456732374,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004984807290209193,
      "loss": 5.3345,
      "mean_token_accuracy": 0.17013389617204666,
      "num_tokens": 9934740.0,
      "step": 5395
    },
    {
      "entropy": 5.54732928276062,
      "epoch": 0.45105245573003677,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004984772733576803,
      "loss": 5.4671,
      "mean_token_accuracy": 0.1761418476700783,
      "num_tokens": 9943516.0,
      "step": 5400
    },
    {
      "entropy": 5.4817307472229,
      "epoch": 0.45147009689274975,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004984738137822028,
      "loss": 5.4043,
      "mean_token_accuracy": 0.16915766149759293,
      "num_tokens": 9952998.0,
      "step": 5405
    },
    {
      "entropy": 5.393400382995606,
      "epoch": 0.45188773805546273,
      "grad_norm": 1.125,
      "learning_rate": 0.000498470350294547,
      "loss": 5.2297,
      "mean_token_accuracy": 0.1843737319111824,
      "num_tokens": 9961698.0,
      "step": 5410
    },
    {
      "entropy": 5.398413372039795,
      "epoch": 0.45230537921817576,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004984668828947739,
      "loss": 5.3753,
      "mean_token_accuracy": 0.1706392526626587,
      "num_tokens": 9972015.0,
      "step": 5415
    },
    {
      "entropy": 5.426323080062867,
      "epoch": 0.45272302038088874,
      "grad_norm": 1.125,
      "learning_rate": 0.0004984634115829439,
      "loss": 5.4487,
      "mean_token_accuracy": 0.1691138908267021,
      "num_tokens": 9981834.0,
      "step": 5420
    },
    {
      "entropy": 5.396098852157593,
      "epoch": 0.4531406615436017,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000498459936359118,
      "loss": 5.3009,
      "mean_token_accuracy": 0.18049199730157853,
      "num_tokens": 9991113.0,
      "step": 5425
    },
    {
      "entropy": 5.384849739074707,
      "epoch": 0.45355830270631475,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004984564572233568,
      "loss": 5.2819,
      "mean_token_accuracy": 0.17862436026334763,
      "num_tokens": 10000048.0,
      "step": 5430
    },
    {
      "entropy": 5.44728536605835,
      "epoch": 0.45397594386902773,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004984529741757216,
      "loss": 5.3259,
      "mean_token_accuracy": 0.1794531986117363,
      "num_tokens": 10009389.0,
      "step": 5435
    },
    {
      "entropy": 5.303836917877197,
      "epoch": 0.4543935850317407,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004984494872162728,
      "loss": 5.1935,
      "mean_token_accuracy": 0.1841413199901581,
      "num_tokens": 10018342.0,
      "step": 5440
    },
    {
      "entropy": 5.4117495059967045,
      "epoch": 0.45481122619445374,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000498445996345072,
      "loss": 5.3622,
      "mean_token_accuracy": 0.17776415944099427,
      "num_tokens": 10027219.0,
      "step": 5445
    },
    {
      "entropy": 5.339814901351929,
      "epoch": 0.4552288673571667,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004984425015621799,
      "loss": 5.2836,
      "mean_token_accuracy": 0.1772462785243988,
      "num_tokens": 10036635.0,
      "step": 5450
    },
    {
      "entropy": 5.326971912384034,
      "epoch": 0.4556465085198797,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000498439002867658,
      "loss": 5.3205,
      "mean_token_accuracy": 0.18105732649564743,
      "num_tokens": 10045655.0,
      "step": 5455
    },
    {
      "entropy": 5.5361474514007565,
      "epoch": 0.45606414968259273,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004984355002615673,
      "loss": 5.3806,
      "mean_token_accuracy": 0.1691015064716339,
      "num_tokens": 10054366.0,
      "step": 5460
    },
    {
      "entropy": 5.434479236602783,
      "epoch": 0.4564817908453057,
      "grad_norm": 1.125,
      "learning_rate": 0.0004984319937439693,
      "loss": 5.3198,
      "mean_token_accuracy": 0.18317292779684066,
      "num_tokens": 10062548.0,
      "step": 5465
    },
    {
      "entropy": 5.3561845302581785,
      "epoch": 0.4568994320080187,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004984284833149252,
      "loss": 5.3105,
      "mean_token_accuracy": 0.17571336179971694,
      "num_tokens": 10071333.0,
      "step": 5470
    },
    {
      "entropy": 5.299349594116211,
      "epoch": 0.4573170731707317,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004984249689744966,
      "loss": 5.2155,
      "mean_token_accuracy": 0.1857534885406494,
      "num_tokens": 10079502.0,
      "step": 5475
    },
    {
      "entropy": 5.36715989112854,
      "epoch": 0.4577347143334447,
      "grad_norm": 1.171875,
      "learning_rate": 0.000498421450722745,
      "loss": 5.2323,
      "mean_token_accuracy": 0.18540998101234435,
      "num_tokens": 10088648.0,
      "step": 5480
    },
    {
      "entropy": 5.363710212707519,
      "epoch": 0.4581523554961577,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000498417928559732,
      "loss": 5.2589,
      "mean_token_accuracy": 0.185733100771904,
      "num_tokens": 10097742.0,
      "step": 5485
    },
    {
      "entropy": 5.437697315216065,
      "epoch": 0.4585699966588707,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004984144024855191,
      "loss": 5.36,
      "mean_token_accuracy": 0.17362416684627532,
      "num_tokens": 10107228.0,
      "step": 5490
    },
    {
      "entropy": 5.4109265327453615,
      "epoch": 0.4589876378215837,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004984108725001683,
      "loss": 5.264,
      "mean_token_accuracy": 0.17965881377458573,
      "num_tokens": 10115488.0,
      "step": 5495
    },
    {
      "entropy": 5.276472520828247,
      "epoch": 0.45940527898429667,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004984073386037413,
      "loss": 5.2027,
      "mean_token_accuracy": 0.18694024235010148,
      "num_tokens": 10124235.0,
      "step": 5500
    },
    {
      "entropy": 5.304714584350586,
      "epoch": 0.4598229201470097,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004984038007962997,
      "loss": 5.3066,
      "mean_token_accuracy": 0.18120416551828383,
      "num_tokens": 10133208.0,
      "step": 5505
    },
    {
      "entropy": 5.3617603302001955,
      "epoch": 0.4602405613097227,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004984002590779059,
      "loss": 5.2291,
      "mean_token_accuracy": 0.17836282253265381,
      "num_tokens": 10142754.0,
      "step": 5510
    },
    {
      "entropy": 5.382237768173217,
      "epoch": 0.46065820247243566,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004983967134486215,
      "loss": 5.3547,
      "mean_token_accuracy": 0.17632970660924913,
      "num_tokens": 10152114.0,
      "step": 5515
    },
    {
      "entropy": 5.41402497291565,
      "epoch": 0.4610758436351487,
      "grad_norm": 1.125,
      "learning_rate": 0.0004983931639085088,
      "loss": 5.2544,
      "mean_token_accuracy": 0.1815406411886215,
      "num_tokens": 10161562.0,
      "step": 5520
    },
    {
      "entropy": 5.416577577590942,
      "epoch": 0.4614934847978617,
      "grad_norm": 1.25,
      "learning_rate": 0.0004983896104576298,
      "loss": 5.3615,
      "mean_token_accuracy": 0.17958488315343857,
      "num_tokens": 10170461.0,
      "step": 5525
    },
    {
      "entropy": 5.388878965377808,
      "epoch": 0.46191112596057465,
      "grad_norm": 1.25,
      "learning_rate": 0.000498386053096047,
      "loss": 5.3639,
      "mean_token_accuracy": 0.17891687750816346,
      "num_tokens": 10179243.0,
      "step": 5530
    },
    {
      "entropy": 5.430717802047729,
      "epoch": 0.4623287671232877,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004983824918238221,
      "loss": 5.3045,
      "mean_token_accuracy": 0.18487516194581985,
      "num_tokens": 10187935.0,
      "step": 5535
    },
    {
      "entropy": 5.411087942123413,
      "epoch": 0.46274640828600067,
      "grad_norm": 1.140625,
      "learning_rate": 0.000498378926641018,
      "loss": 5.377,
      "mean_token_accuracy": 0.17247853726148604,
      "num_tokens": 10196185.0,
      "step": 5540
    },
    {
      "entropy": 5.401890468597412,
      "epoch": 0.46316404944871364,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004983753575476969,
      "loss": 5.3664,
      "mean_token_accuracy": 0.17471231073141097,
      "num_tokens": 10205373.0,
      "step": 5545
    },
    {
      "entropy": 5.46509485244751,
      "epoch": 0.4635816906114267,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004983717845439212,
      "loss": 5.3357,
      "mean_token_accuracy": 0.18027413636446,
      "num_tokens": 10214464.0,
      "step": 5550
    },
    {
      "entropy": 5.4192205429077145,
      "epoch": 0.46399933177413966,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004983682076297536,
      "loss": 5.4152,
      "mean_token_accuracy": 0.17256004810333253,
      "num_tokens": 10224462.0,
      "step": 5555
    },
    {
      "entropy": 5.460912370681763,
      "epoch": 0.46441697293685263,
      "grad_norm": 1.25,
      "learning_rate": 0.0004983646268052567,
      "loss": 5.3335,
      "mean_token_accuracy": 0.17892708480358124,
      "num_tokens": 10233123.0,
      "step": 5560
    },
    {
      "entropy": 5.3775170803070065,
      "epoch": 0.46483461409956567,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004983610420704931,
      "loss": 5.1826,
      "mean_token_accuracy": 0.18358879983425141,
      "num_tokens": 10241613.0,
      "step": 5565
    },
    {
      "entropy": 5.43302698135376,
      "epoch": 0.46525225526227865,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004983574534255256,
      "loss": 5.5336,
      "mean_token_accuracy": 0.16362907588481904,
      "num_tokens": 10251809.0,
      "step": 5570
    },
    {
      "entropy": 5.384333086013794,
      "epoch": 0.4656698964249916,
      "grad_norm": 1.296875,
      "learning_rate": 0.000498353860870417,
      "loss": 5.2102,
      "mean_token_accuracy": 0.17986831367015838,
      "num_tokens": 10260198.0,
      "step": 5575
    },
    {
      "entropy": 5.413422966003418,
      "epoch": 0.46608753758770466,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004983502644052303,
      "loss": 5.2964,
      "mean_token_accuracy": 0.18045338243246078,
      "num_tokens": 10268906.0,
      "step": 5580
    },
    {
      "entropy": 5.293640184402466,
      "epoch": 0.46650517875041764,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004983466640300282,
      "loss": 5.1946,
      "mean_token_accuracy": 0.17398974746465684,
      "num_tokens": 10278526.0,
      "step": 5585
    },
    {
      "entropy": 5.355238723754883,
      "epoch": 0.4669228199131306,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004983430597448741,
      "loss": 5.313,
      "mean_token_accuracy": 0.17971521466970444,
      "num_tokens": 10287428.0,
      "step": 5590
    },
    {
      "entropy": 5.497137498855591,
      "epoch": 0.46734046107584365,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004983394515498308,
      "loss": 5.303,
      "mean_token_accuracy": 0.17363718301057815,
      "num_tokens": 10297136.0,
      "step": 5595
    },
    {
      "entropy": 5.3973921775817875,
      "epoch": 0.46775810223855663,
      "grad_norm": 1.125,
      "learning_rate": 0.0004983358394449616,
      "loss": 5.3768,
      "mean_token_accuracy": 0.17228453010320663,
      "num_tokens": 10305678.0,
      "step": 5600
    },
    {
      "entropy": 5.335638189315796,
      "epoch": 0.4681757434012696,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004983322234303297,
      "loss": 5.2924,
      "mean_token_accuracy": 0.18027922958135606,
      "num_tokens": 10314690.0,
      "step": 5605
    },
    {
      "entropy": 5.358850812911987,
      "epoch": 0.46859338456398264,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004983286035059984,
      "loss": 5.2356,
      "mean_token_accuracy": 0.1823755368590355,
      "num_tokens": 10323521.0,
      "step": 5610
    },
    {
      "entropy": 5.38775372505188,
      "epoch": 0.4690110257266956,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004983249796720311,
      "loss": 5.2839,
      "mean_token_accuracy": 0.18415401577949525,
      "num_tokens": 10333077.0,
      "step": 5615
    },
    {
      "entropy": 5.395961093902588,
      "epoch": 0.4694286668894086,
      "grad_norm": 1.625,
      "learning_rate": 0.0004983213519284912,
      "loss": 5.3284,
      "mean_token_accuracy": 0.17712220549583435,
      "num_tokens": 10342194.0,
      "step": 5620
    },
    {
      "entropy": 5.370408296585083,
      "epoch": 0.46984630805212163,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004983177202754421,
      "loss": 5.1725,
      "mean_token_accuracy": 0.18557369709014893,
      "num_tokens": 10351009.0,
      "step": 5625
    },
    {
      "entropy": 5.418461847305298,
      "epoch": 0.4702639492148346,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004983140847129476,
      "loss": 5.3652,
      "mean_token_accuracy": 0.1745539128780365,
      "num_tokens": 10360872.0,
      "step": 5630
    },
    {
      "entropy": 5.362072467803955,
      "epoch": 0.4706815903775476,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004983104452410713,
      "loss": 5.2709,
      "mean_token_accuracy": 0.18330505192279817,
      "num_tokens": 10370159.0,
      "step": 5635
    },
    {
      "entropy": 5.361561489105225,
      "epoch": 0.4710992315402606,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004983068018598768,
      "loss": 5.277,
      "mean_token_accuracy": 0.1826947033405304,
      "num_tokens": 10379219.0,
      "step": 5640
    },
    {
      "entropy": 5.415415096282959,
      "epoch": 0.4715168727029736,
      "grad_norm": 1.171875,
      "learning_rate": 0.000498303154569428,
      "loss": 5.3165,
      "mean_token_accuracy": 0.17777066081762313,
      "num_tokens": 10389425.0,
      "step": 5645
    },
    {
      "entropy": 5.404468965530396,
      "epoch": 0.4719345138656866,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004982995033697887,
      "loss": 5.3426,
      "mean_token_accuracy": 0.1823611155152321,
      "num_tokens": 10398023.0,
      "step": 5650
    },
    {
      "entropy": 5.3171977519989015,
      "epoch": 0.4723521550283996,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004982958482610226,
      "loss": 5.2605,
      "mean_token_accuracy": 0.18108674585819245,
      "num_tokens": 10406578.0,
      "step": 5655
    },
    {
      "entropy": 5.324284410476684,
      "epoch": 0.4727697961911126,
      "grad_norm": 1.125,
      "learning_rate": 0.000498292189243194,
      "loss": 5.2865,
      "mean_token_accuracy": 0.18560549318790437,
      "num_tokens": 10416448.0,
      "step": 5660
    },
    {
      "entropy": 5.442724084854126,
      "epoch": 0.47318743735382557,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000498288526316367,
      "loss": 5.301,
      "mean_token_accuracy": 0.17472774982452394,
      "num_tokens": 10425414.0,
      "step": 5665
    },
    {
      "entropy": 5.3667394638061525,
      "epoch": 0.4736050785165386,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004982848594806054,
      "loss": 5.2899,
      "mean_token_accuracy": 0.18441671282052993,
      "num_tokens": 10433900.0,
      "step": 5670
    },
    {
      "entropy": 5.378509855270385,
      "epoch": 0.4740227196792516,
      "grad_norm": 1.25,
      "learning_rate": 0.0004982811887359736,
      "loss": 5.4014,
      "mean_token_accuracy": 0.17403701096773147,
      "num_tokens": 10443077.0,
      "step": 5675
    },
    {
      "entropy": 5.416761207580566,
      "epoch": 0.47444036084196456,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004982775140825359,
      "loss": 5.3106,
      "mean_token_accuracy": 0.1761450931429863,
      "num_tokens": 10452791.0,
      "step": 5680
    },
    {
      "entropy": 5.496857118606568,
      "epoch": 0.4748580020046776,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004982738355203565,
      "loss": 5.4832,
      "mean_token_accuracy": 0.17297002524137498,
      "num_tokens": 10462087.0,
      "step": 5685
    },
    {
      "entropy": 5.523170042037964,
      "epoch": 0.47527564316739057,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004982701530494998,
      "loss": 5.5053,
      "mean_token_accuracy": 0.16588604748249053,
      "num_tokens": 10472576.0,
      "step": 5690
    },
    {
      "entropy": 5.419825744628906,
      "epoch": 0.47569328433010355,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004982664666700304,
      "loss": 5.219,
      "mean_token_accuracy": 0.18905199021101,
      "num_tokens": 10481176.0,
      "step": 5695
    },
    {
      "entropy": 5.340116405487061,
      "epoch": 0.4761109254928166,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004982627763820128,
      "loss": 5.2996,
      "mean_token_accuracy": 0.179258593916893,
      "num_tokens": 10490558.0,
      "step": 5700
    },
    {
      "entropy": 5.399360990524292,
      "epoch": 0.47652856665552956,
      "grad_norm": 1.125,
      "learning_rate": 0.0004982590821855115,
      "loss": 5.3737,
      "mean_token_accuracy": 0.17810954302549362,
      "num_tokens": 10499543.0,
      "step": 5705
    },
    {
      "entropy": 5.3109029769897464,
      "epoch": 0.47694620781824254,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004982553840805912,
      "loss": 5.2672,
      "mean_token_accuracy": 0.183368119597435,
      "num_tokens": 10508162.0,
      "step": 5710
    },
    {
      "entropy": 5.337470769882202,
      "epoch": 0.4773638489809556,
      "grad_norm": 1.25,
      "learning_rate": 0.0004982516820673167,
      "loss": 5.1699,
      "mean_token_accuracy": 0.18151225298643112,
      "num_tokens": 10516837.0,
      "step": 5715
    },
    {
      "entropy": 5.3128574848175045,
      "epoch": 0.47778149014366855,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004982479761457528,
      "loss": 5.3232,
      "mean_token_accuracy": 0.17841623425483705,
      "num_tokens": 10526208.0,
      "step": 5720
    },
    {
      "entropy": 5.42949161529541,
      "epoch": 0.47819913130638153,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004982442663159643,
      "loss": 5.3616,
      "mean_token_accuracy": 0.1749148890376091,
      "num_tokens": 10536002.0,
      "step": 5725
    },
    {
      "entropy": 5.465812397003174,
      "epoch": 0.47861677246909456,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004982405525780162,
      "loss": 5.3403,
      "mean_token_accuracy": 0.17801699340343474,
      "num_tokens": 10545861.0,
      "step": 5730
    },
    {
      "entropy": 5.3826446533203125,
      "epoch": 0.47903441363180754,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004982368349319736,
      "loss": 5.2491,
      "mean_token_accuracy": 0.1851024940609932,
      "num_tokens": 10555357.0,
      "step": 5735
    },
    {
      "entropy": 5.3701520442962645,
      "epoch": 0.4794520547945205,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004982331133779014,
      "loss": 5.3228,
      "mean_token_accuracy": 0.1765170603990555,
      "num_tokens": 10564744.0,
      "step": 5740
    },
    {
      "entropy": 5.355828952789307,
      "epoch": 0.47986969595723356,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004982293879158649,
      "loss": 5.2753,
      "mean_token_accuracy": 0.18183653056621552,
      "num_tokens": 10573736.0,
      "step": 5745
    },
    {
      "entropy": 5.376202154159546,
      "epoch": 0.48028733711994653,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004982256585459293,
      "loss": 5.3591,
      "mean_token_accuracy": 0.17756519019603728,
      "num_tokens": 10583238.0,
      "step": 5750
    },
    {
      "entropy": 5.462325572967529,
      "epoch": 0.4807049782826595,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004982219252681597,
      "loss": 5.3997,
      "mean_token_accuracy": 0.17141326367855073,
      "num_tokens": 10592316.0,
      "step": 5755
    },
    {
      "entropy": 5.41369743347168,
      "epoch": 0.48112261944537255,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004982181880826218,
      "loss": 5.421,
      "mean_token_accuracy": 0.177415369451046,
      "num_tokens": 10601418.0,
      "step": 5760
    },
    {
      "entropy": 5.3490503311157225,
      "epoch": 0.4815402606080855,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004982144469893807,
      "loss": 5.2801,
      "mean_token_accuracy": 0.18411952406167983,
      "num_tokens": 10610513.0,
      "step": 5765
    },
    {
      "entropy": 5.408094787597657,
      "epoch": 0.4819579017707985,
      "grad_norm": 1.109375,
      "learning_rate": 0.000498210701988502,
      "loss": 5.2443,
      "mean_token_accuracy": 0.18783568739891052,
      "num_tokens": 10619529.0,
      "step": 5770
    },
    {
      "entropy": 5.453507995605468,
      "epoch": 0.48237554293351154,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004982069530800514,
      "loss": 5.3288,
      "mean_token_accuracy": 0.18438346534967423,
      "num_tokens": 10628263.0,
      "step": 5775
    },
    {
      "entropy": 5.3486114025115965,
      "epoch": 0.4827931840962245,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004982032002640943,
      "loss": 5.3463,
      "mean_token_accuracy": 0.1833382233977318,
      "num_tokens": 10637605.0,
      "step": 5780
    },
    {
      "entropy": 5.416544818878174,
      "epoch": 0.48321082525893755,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004981994435406965,
      "loss": 5.4228,
      "mean_token_accuracy": 0.17957475930452346,
      "num_tokens": 10647113.0,
      "step": 5785
    },
    {
      "entropy": 5.40274977684021,
      "epoch": 0.4836284664216505,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004981956829099239,
      "loss": 5.1838,
      "mean_token_accuracy": 0.18768427073955535,
      "num_tokens": 10655464.0,
      "step": 5790
    },
    {
      "entropy": 5.359679365158081,
      "epoch": 0.4840461075843635,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004981919183718422,
      "loss": 5.2834,
      "mean_token_accuracy": 0.18101781010627746,
      "num_tokens": 10664689.0,
      "step": 5795
    },
    {
      "entropy": 5.2755897521972654,
      "epoch": 0.48446374874707654,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004981881499265173,
      "loss": 5.201,
      "mean_token_accuracy": 0.18344079703092575,
      "num_tokens": 10673440.0,
      "step": 5800
    },
    {
      "entropy": 5.342781734466553,
      "epoch": 0.4848813899097895,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004981843775740151,
      "loss": 5.3392,
      "mean_token_accuracy": 0.1783262863755226,
      "num_tokens": 10683075.0,
      "step": 5805
    },
    {
      "entropy": 5.365957736968994,
      "epoch": 0.4852990310725025,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004981806013144017,
      "loss": 5.2949,
      "mean_token_accuracy": 0.18161205500364302,
      "num_tokens": 10692663.0,
      "step": 5810
    },
    {
      "entropy": 5.356691837310791,
      "epoch": 0.48571667223521553,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004981768211477434,
      "loss": 5.1803,
      "mean_token_accuracy": 0.18302590399980545,
      "num_tokens": 10702068.0,
      "step": 5815
    },
    {
      "entropy": 5.388083744049072,
      "epoch": 0.4861343133979285,
      "grad_norm": 1.140625,
      "learning_rate": 0.000498173037074106,
      "loss": 5.3313,
      "mean_token_accuracy": 0.1714998409152031,
      "num_tokens": 10711230.0,
      "step": 5820
    },
    {
      "entropy": 5.2951349258422855,
      "epoch": 0.4865519545606415,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004981692490935561,
      "loss": 5.2956,
      "mean_token_accuracy": 0.18595292121171952,
      "num_tokens": 10719651.0,
      "step": 5825
    },
    {
      "entropy": 5.396548748016357,
      "epoch": 0.4869695957233545,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004981654572061597,
      "loss": 5.3138,
      "mean_token_accuracy": 0.17995359599590302,
      "num_tokens": 10729342.0,
      "step": 5830
    },
    {
      "entropy": 5.432379722595215,
      "epoch": 0.4873872368860675,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004981616614119836,
      "loss": 5.3094,
      "mean_token_accuracy": 0.18182204365730287,
      "num_tokens": 10738617.0,
      "step": 5835
    },
    {
      "entropy": 5.370879411697388,
      "epoch": 0.4878048780487805,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004981578617110938,
      "loss": 5.352,
      "mean_token_accuracy": 0.1767614170908928,
      "num_tokens": 10746881.0,
      "step": 5840
    },
    {
      "entropy": 5.33872241973877,
      "epoch": 0.4882225192114935,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004981540581035571,
      "loss": 5.2133,
      "mean_token_accuracy": 0.18773332238197327,
      "num_tokens": 10755848.0,
      "step": 5845
    },
    {
      "entropy": 5.292233085632324,
      "epoch": 0.4886401603742065,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00049815025058944,
      "loss": 5.2107,
      "mean_token_accuracy": 0.17764829993247985,
      "num_tokens": 10765519.0,
      "step": 5850
    },
    {
      "entropy": 5.328621339797974,
      "epoch": 0.48905780153691947,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004981464391688092,
      "loss": 5.3449,
      "mean_token_accuracy": 0.17469416856765746,
      "num_tokens": 10776408.0,
      "step": 5855
    },
    {
      "entropy": 5.430347108840943,
      "epoch": 0.4894754426996325,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004981426238417312,
      "loss": 5.3232,
      "mean_token_accuracy": 0.18280997276306152,
      "num_tokens": 10785499.0,
      "step": 5860
    },
    {
      "entropy": 5.363692855834961,
      "epoch": 0.4898930838623455,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004981388046082731,
      "loss": 5.2791,
      "mean_token_accuracy": 0.17865316569805145,
      "num_tokens": 10794736.0,
      "step": 5865
    },
    {
      "entropy": 5.346862268447876,
      "epoch": 0.49031072502505846,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004981349814685016,
      "loss": 5.3522,
      "mean_token_accuracy": 0.17651747465133666,
      "num_tokens": 10804125.0,
      "step": 5870
    },
    {
      "entropy": 5.457276105880737,
      "epoch": 0.4907283661877715,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004981311544224838,
      "loss": 5.3986,
      "mean_token_accuracy": 0.17904023975133895,
      "num_tokens": 10813465.0,
      "step": 5875
    },
    {
      "entropy": 5.357408809661865,
      "epoch": 0.49114600735048447,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004981273234702864,
      "loss": 5.2211,
      "mean_token_accuracy": 0.1853638082742691,
      "num_tokens": 10821550.0,
      "step": 5880
    },
    {
      "entropy": 5.320731067657471,
      "epoch": 0.49156364851319745,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004981234886119767,
      "loss": 5.2485,
      "mean_token_accuracy": 0.18118553310632707,
      "num_tokens": 10830683.0,
      "step": 5885
    },
    {
      "entropy": 5.368787336349487,
      "epoch": 0.4919812896759105,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004981196498476217,
      "loss": 5.272,
      "mean_token_accuracy": 0.18537646234035493,
      "num_tokens": 10839492.0,
      "step": 5890
    },
    {
      "entropy": 5.318939924240112,
      "epoch": 0.49239893083862346,
      "grad_norm": 1.25,
      "learning_rate": 0.0004981158071772887,
      "loss": 5.1956,
      "mean_token_accuracy": 0.18439319729804993,
      "num_tokens": 10847998.0,
      "step": 5895
    },
    {
      "entropy": 5.235811233520508,
      "epoch": 0.49281657200133644,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004981119606010449,
      "loss": 5.2019,
      "mean_token_accuracy": 0.1859842360019684,
      "num_tokens": 10856438.0,
      "step": 5900
    },
    {
      "entropy": 5.2461645126342775,
      "epoch": 0.4932342131640495,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004981081101189577,
      "loss": 5.2036,
      "mean_token_accuracy": 0.18382869511842728,
      "num_tokens": 10865936.0,
      "step": 5905
    },
    {
      "entropy": 5.419135046005249,
      "epoch": 0.49365185432676245,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004981042557310945,
      "loss": 5.2117,
      "mean_token_accuracy": 0.18635481595993042,
      "num_tokens": 10874621.0,
      "step": 5910
    },
    {
      "entropy": 5.294868326187133,
      "epoch": 0.49406949548947543,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004981003974375227,
      "loss": 5.179,
      "mean_token_accuracy": 0.18442256897687911,
      "num_tokens": 10883516.0,
      "step": 5915
    },
    {
      "entropy": 5.328514528274536,
      "epoch": 0.49448713665218846,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004980965352383099,
      "loss": 5.3253,
      "mean_token_accuracy": 0.1761163741350174,
      "num_tokens": 10892580.0,
      "step": 5920
    },
    {
      "entropy": 5.409572649002075,
      "epoch": 0.49490477781490144,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004980926691335238,
      "loss": 5.2566,
      "mean_token_accuracy": 0.18117301613092424,
      "num_tokens": 10901227.0,
      "step": 5925
    },
    {
      "entropy": 5.367732334136963,
      "epoch": 0.4953224189776144,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004980887991232318,
      "loss": 5.2989,
      "mean_token_accuracy": 0.17467935681343078,
      "num_tokens": 10909454.0,
      "step": 5930
    },
    {
      "entropy": 5.482755661010742,
      "epoch": 0.49574006014032745,
      "grad_norm": 1.203125,
      "learning_rate": 0.000498084925207502,
      "loss": 5.3645,
      "mean_token_accuracy": 0.1773775488138199,
      "num_tokens": 10919792.0,
      "step": 5935
    },
    {
      "entropy": 5.419538593292236,
      "epoch": 0.49615770130304043,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000498081047386402,
      "loss": 5.4197,
      "mean_token_accuracy": 0.17599386274814605,
      "num_tokens": 10930523.0,
      "step": 5940
    },
    {
      "entropy": 5.309274291992187,
      "epoch": 0.4965753424657534,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004980771656599998,
      "loss": 5.263,
      "mean_token_accuracy": 0.1857204556465149,
      "num_tokens": 10940075.0,
      "step": 5945
    },
    {
      "entropy": 5.3744665622711185,
      "epoch": 0.49699298362846644,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004980732800283633,
      "loss": 5.2694,
      "mean_token_accuracy": 0.18188395649194716,
      "num_tokens": 10949059.0,
      "step": 5950
    },
    {
      "entropy": 5.357154750823975,
      "epoch": 0.4974106247911794,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004980693904915604,
      "loss": 5.2599,
      "mean_token_accuracy": 0.18225459307432174,
      "num_tokens": 10958313.0,
      "step": 5955
    },
    {
      "entropy": 5.327280139923095,
      "epoch": 0.4978282659538924,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004980654970496595,
      "loss": 5.3083,
      "mean_token_accuracy": 0.17470995038747789,
      "num_tokens": 10967666.0,
      "step": 5960
    },
    {
      "entropy": 5.457955408096313,
      "epoch": 0.49824590711660544,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004980615997027284,
      "loss": 5.3324,
      "mean_token_accuracy": 0.18239685744047165,
      "num_tokens": 10976519.0,
      "step": 5965
    },
    {
      "entropy": 5.47369351387024,
      "epoch": 0.4986635482793184,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004980576984508356,
      "loss": 5.4425,
      "mean_token_accuracy": 0.16856858879327774,
      "num_tokens": 10987138.0,
      "step": 5970
    },
    {
      "entropy": 5.424091768264771,
      "epoch": 0.4990811894420314,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004980537932940492,
      "loss": 5.3092,
      "mean_token_accuracy": 0.18266766965389253,
      "num_tokens": 10996517.0,
      "step": 5975
    },
    {
      "entropy": 5.445526170730591,
      "epoch": 0.4994988306047444,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004980498842324377,
      "loss": 5.1936,
      "mean_token_accuracy": 0.17778559327125548,
      "num_tokens": 11006312.0,
      "step": 5980
    },
    {
      "entropy": 5.303483533859253,
      "epoch": 0.4999164717674574,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004980459712660694,
      "loss": 5.2982,
      "mean_token_accuracy": 0.1875140354037285,
      "num_tokens": 11015535.0,
      "step": 5985
    },
    {
      "entropy": 5.36759123802185,
      "epoch": 0.5003341129301704,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004980420543950131,
      "loss": 5.3104,
      "mean_token_accuracy": 0.18209726810455323,
      "num_tokens": 11024887.0,
      "step": 5990
    },
    {
      "entropy": 5.2870267868042,
      "epoch": 0.5007517540928834,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004980381336193371,
      "loss": 5.1645,
      "mean_token_accuracy": 0.19015818387269973,
      "num_tokens": 11033147.0,
      "step": 5995
    },
    {
      "entropy": 5.295065307617188,
      "epoch": 0.5011693952555963,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004980342089391099,
      "loss": 5.3273,
      "mean_token_accuracy": 0.1844463363289833,
      "num_tokens": 11042726.0,
      "step": 6000
    },
    {
      "epoch": 0.5011693952555963,
      "eval_entropy": 5.172037276037919,
      "eval_loss": 5.286351680755615,
      "eval_mean_token_accuracy": 0.19003585199226744,
      "eval_num_tokens": 11042726.0,
      "eval_runtime": 27.6328,
      "eval_samples_per_second": 1366.6,
      "eval_steps_per_second": 170.848,
      "step": 6000
    },
    {
      "entropy": 5.414085102081299,
      "epoch": 0.5015870364183094,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004980302803544006,
      "loss": 5.2927,
      "mean_token_accuracy": 0.18288459330797197,
      "num_tokens": 11052181.0,
      "step": 6005
    },
    {
      "entropy": 5.431057929992676,
      "epoch": 0.5020046775810224,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004980263478652778,
      "loss": 5.3376,
      "mean_token_accuracy": 0.1821412682533264,
      "num_tokens": 11061586.0,
      "step": 6010
    },
    {
      "entropy": 5.46299524307251,
      "epoch": 0.5024223187437353,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004980224114718102,
      "loss": 5.3663,
      "mean_token_accuracy": 0.18216219246387483,
      "num_tokens": 11070914.0,
      "step": 6015
    },
    {
      "entropy": 5.320691680908203,
      "epoch": 0.5028399599064484,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004980184711740669,
      "loss": 5.1881,
      "mean_token_accuracy": 0.1889343485236168,
      "num_tokens": 11079836.0,
      "step": 6020
    },
    {
      "entropy": 5.273727035522461,
      "epoch": 0.5032576010691614,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004980145269721168,
      "loss": 5.1989,
      "mean_token_accuracy": 0.18011632412672043,
      "num_tokens": 11088320.0,
      "step": 6025
    },
    {
      "entropy": 5.405125617980957,
      "epoch": 0.5036752422318743,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000498010578866029,
      "loss": 5.3731,
      "mean_token_accuracy": 0.1793920487165451,
      "num_tokens": 11097803.0,
      "step": 6030
    },
    {
      "entropy": 5.513732242584228,
      "epoch": 0.5040928833945874,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004980066268558726,
      "loss": 5.3942,
      "mean_token_accuracy": 0.17209561616182328,
      "num_tokens": 11107218.0,
      "step": 6035
    },
    {
      "entropy": 5.453762054443359,
      "epoch": 0.5045105245573004,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004980026709417167,
      "loss": 5.2838,
      "mean_token_accuracy": 0.17526641935110093,
      "num_tokens": 11115599.0,
      "step": 6040
    },
    {
      "entropy": 5.254436206817627,
      "epoch": 0.5049281657200133,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004979987111236306,
      "loss": 5.1495,
      "mean_token_accuracy": 0.1871658131480217,
      "num_tokens": 11124312.0,
      "step": 6045
    },
    {
      "entropy": 5.320400142669678,
      "epoch": 0.5053458068827263,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004979947474016837,
      "loss": 5.2926,
      "mean_token_accuracy": 0.18224723339080812,
      "num_tokens": 11133117.0,
      "step": 6050
    },
    {
      "entropy": 5.485568237304688,
      "epoch": 0.5057634480454394,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004979907797759453,
      "loss": 5.3359,
      "mean_token_accuracy": 0.18229107707738876,
      "num_tokens": 11141972.0,
      "step": 6055
    },
    {
      "entropy": 5.489246559143067,
      "epoch": 0.5061810892081523,
      "grad_norm": 1.125,
      "learning_rate": 0.0004979868082464849,
      "loss": 5.3555,
      "mean_token_accuracy": 0.17801129072904587,
      "num_tokens": 11151416.0,
      "step": 6060
    },
    {
      "entropy": 5.368074750900268,
      "epoch": 0.5065987303708653,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004979828328133719,
      "loss": 5.3028,
      "mean_token_accuracy": 0.1798588827252388,
      "num_tokens": 11160727.0,
      "step": 6065
    },
    {
      "entropy": 5.34806489944458,
      "epoch": 0.5070163715335784,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004979788534766762,
      "loss": 5.2715,
      "mean_token_accuracy": 0.17353069931268691,
      "num_tokens": 11171350.0,
      "step": 6070
    },
    {
      "entropy": 5.359587049484253,
      "epoch": 0.5074340126962913,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004979748702364672,
      "loss": 5.248,
      "mean_token_accuracy": 0.18068615347146988,
      "num_tokens": 11180233.0,
      "step": 6075
    },
    {
      "entropy": 5.349366998672485,
      "epoch": 0.5078516538590043,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004979708830928147,
      "loss": 5.2477,
      "mean_token_accuracy": 0.17650825530290604,
      "num_tokens": 11189269.0,
      "step": 6080
    },
    {
      "entropy": 5.3739677429199215,
      "epoch": 0.5082692950217174,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004979668920457884,
      "loss": 5.3131,
      "mean_token_accuracy": 0.1810814157128334,
      "num_tokens": 11199072.0,
      "step": 6085
    },
    {
      "entropy": 5.3011698722839355,
      "epoch": 0.5086869361844303,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004979628970954584,
      "loss": 5.212,
      "mean_token_accuracy": 0.18820451498031615,
      "num_tokens": 11207164.0,
      "step": 6090
    },
    {
      "entropy": 5.339222431182861,
      "epoch": 0.5091045773471433,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004979588982418944,
      "loss": 5.2644,
      "mean_token_accuracy": 0.17381262183189392,
      "num_tokens": 11216557.0,
      "step": 6095
    },
    {
      "entropy": 5.52068681716919,
      "epoch": 0.5095222185098564,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004979548954851666,
      "loss": 5.4143,
      "mean_token_accuracy": 0.1751055583357811,
      "num_tokens": 11227277.0,
      "step": 6100
    },
    {
      "entropy": 5.401752376556397,
      "epoch": 0.5099398596725693,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000497950888825345,
      "loss": 5.2005,
      "mean_token_accuracy": 0.19273057878017424,
      "num_tokens": 11236226.0,
      "step": 6105
    },
    {
      "entropy": 5.343244981765747,
      "epoch": 0.5103575008352823,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004979468782624996,
      "loss": 5.3121,
      "mean_token_accuracy": 0.18312968611717223,
      "num_tokens": 11245246.0,
      "step": 6110
    },
    {
      "entropy": 5.366712760925293,
      "epoch": 0.5107751419979953,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004979428637967009,
      "loss": 5.372,
      "mean_token_accuracy": 0.1794750526547432,
      "num_tokens": 11254437.0,
      "step": 6115
    },
    {
      "entropy": 5.381765365600586,
      "epoch": 0.5111927831607083,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004979388454280189,
      "loss": 5.2136,
      "mean_token_accuracy": 0.18831336945295335,
      "num_tokens": 11262413.0,
      "step": 6120
    },
    {
      "entropy": 5.3759208679199215,
      "epoch": 0.5116104243234213,
      "grad_norm": 1.359375,
      "learning_rate": 0.000497934823156524,
      "loss": 5.4007,
      "mean_token_accuracy": 0.17576018571853638,
      "num_tokens": 11272061.0,
      "step": 6125
    },
    {
      "entropy": 5.353822851181031,
      "epoch": 0.5120280654861343,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004979307969822869,
      "loss": 5.2244,
      "mean_token_accuracy": 0.1847929373383522,
      "num_tokens": 11281623.0,
      "step": 6130
    },
    {
      "entropy": 5.2704144477844235,
      "epoch": 0.5124457066488474,
      "grad_norm": 1.25,
      "learning_rate": 0.0004979267669053776,
      "loss": 5.1026,
      "mean_token_accuracy": 0.1965145394206047,
      "num_tokens": 11290202.0,
      "step": 6135
    },
    {
      "entropy": 5.346707725524903,
      "epoch": 0.5128633478115603,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004979227329258671,
      "loss": 5.3702,
      "mean_token_accuracy": 0.18677955716848374,
      "num_tokens": 11299040.0,
      "step": 6140
    },
    {
      "entropy": 5.4291024684906,
      "epoch": 0.5132809889742733,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004979186950438256,
      "loss": 5.295,
      "mean_token_accuracy": 0.17737863659858705,
      "num_tokens": 11307938.0,
      "step": 6145
    },
    {
      "entropy": 5.417812776565552,
      "epoch": 0.5136986301369864,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004979146532593243,
      "loss": 5.2973,
      "mean_token_accuracy": 0.18085166364908217,
      "num_tokens": 11316915.0,
      "step": 6150
    },
    {
      "entropy": 5.36984429359436,
      "epoch": 0.5141162712996993,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004979106075724334,
      "loss": 5.2635,
      "mean_token_accuracy": 0.1764444351196289,
      "num_tokens": 11326634.0,
      "step": 6155
    },
    {
      "entropy": 5.471321630477905,
      "epoch": 0.5145339124624123,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004979065579832242,
      "loss": 5.4438,
      "mean_token_accuracy": 0.16635178923606872,
      "num_tokens": 11336502.0,
      "step": 6160
    },
    {
      "entropy": 5.374242877960205,
      "epoch": 0.5149515536251253,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004979025044917673,
      "loss": 5.3295,
      "mean_token_accuracy": 0.17512160092592238,
      "num_tokens": 11346125.0,
      "step": 6165
    },
    {
      "entropy": 5.439322662353516,
      "epoch": 0.5153691947878383,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004978984470981339,
      "loss": 5.4209,
      "mean_token_accuracy": 0.16800944954156877,
      "num_tokens": 11355713.0,
      "step": 6170
    },
    {
      "entropy": 5.390084266662598,
      "epoch": 0.5157868359505513,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004978943858023946,
      "loss": 5.2302,
      "mean_token_accuracy": 0.18250556588172911,
      "num_tokens": 11364726.0,
      "step": 6175
    },
    {
      "entropy": 5.398955154418945,
      "epoch": 0.5162044771132643,
      "grad_norm": 1.21875,
      "learning_rate": 0.000497890320604621,
      "loss": 5.2336,
      "mean_token_accuracy": 0.1793229892849922,
      "num_tokens": 11373779.0,
      "step": 6180
    },
    {
      "entropy": 5.286739683151245,
      "epoch": 0.5166221182759773,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000497886251504884,
      "loss": 5.2421,
      "mean_token_accuracy": 0.1845567375421524,
      "num_tokens": 11382641.0,
      "step": 6185
    },
    {
      "entropy": 5.2926445484161375,
      "epoch": 0.5170397594386903,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004978821785032549,
      "loss": 5.21,
      "mean_token_accuracy": 0.1850380316376686,
      "num_tokens": 11392200.0,
      "step": 6190
    },
    {
      "entropy": 5.4064432144165036,
      "epoch": 0.5174574006014033,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004978781015998048,
      "loss": 5.3277,
      "mean_token_accuracy": 0.16943650394678117,
      "num_tokens": 11401884.0,
      "step": 6195
    },
    {
      "entropy": 5.344465208053589,
      "epoch": 0.5178750417641163,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004978740207946056,
      "loss": 5.2197,
      "mean_token_accuracy": 0.1897432804107666,
      "num_tokens": 11410710.0,
      "step": 6200
    },
    {
      "entropy": 5.275054216384888,
      "epoch": 0.5182926829268293,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004978699360877281,
      "loss": 5.1735,
      "mean_token_accuracy": 0.18877318501472473,
      "num_tokens": 11419622.0,
      "step": 6205
    },
    {
      "entropy": 5.327790784835815,
      "epoch": 0.5187103240895423,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004978658474792442,
      "loss": 5.1913,
      "mean_token_accuracy": 0.1880590006709099,
      "num_tokens": 11427631.0,
      "step": 6210
    },
    {
      "entropy": 5.373764514923096,
      "epoch": 0.5191279652522552,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004978617549692253,
      "loss": 5.249,
      "mean_token_accuracy": 0.18507532477378846,
      "num_tokens": 11437630.0,
      "step": 6215
    },
    {
      "entropy": 5.479686069488525,
      "epoch": 0.5195456064149683,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004978576585577432,
      "loss": 5.4764,
      "mean_token_accuracy": 0.16511390805244447,
      "num_tokens": 11447962.0,
      "step": 6220
    },
    {
      "entropy": 5.501290321350098,
      "epoch": 0.5199632475776813,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004978535582448695,
      "loss": 5.3293,
      "mean_token_accuracy": 0.1783910036087036,
      "num_tokens": 11458154.0,
      "step": 6225
    },
    {
      "entropy": 5.261790037155151,
      "epoch": 0.5203808887403942,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004978494540306761,
      "loss": 5.1856,
      "mean_token_accuracy": 0.18092765361070634,
      "num_tokens": 11467449.0,
      "step": 6230
    },
    {
      "entropy": 5.3864082336425785,
      "epoch": 0.5207985299031073,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004978453459152346,
      "loss": 5.4549,
      "mean_token_accuracy": 0.1697009190917015,
      "num_tokens": 11477775.0,
      "step": 6235
    },
    {
      "entropy": 5.486942911148072,
      "epoch": 0.5212161710658203,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004978412338986172,
      "loss": 5.2283,
      "mean_token_accuracy": 0.17945119887590408,
      "num_tokens": 11486921.0,
      "step": 6240
    },
    {
      "entropy": 5.283312463760376,
      "epoch": 0.5216338122285332,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004978371179808958,
      "loss": 5.2413,
      "mean_token_accuracy": 0.18170327991247176,
      "num_tokens": 11496562.0,
      "step": 6245
    },
    {
      "entropy": 5.331500291824341,
      "epoch": 0.5220514533912463,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004978329981621424,
      "loss": 5.1934,
      "mean_token_accuracy": 0.18204472064971924,
      "num_tokens": 11504906.0,
      "step": 6250
    },
    {
      "entropy": 5.365571546554565,
      "epoch": 0.5224690945539593,
      "grad_norm": 1.625,
      "learning_rate": 0.000497828874442429,
      "loss": 5.2457,
      "mean_token_accuracy": 0.18381647765636444,
      "num_tokens": 11513757.0,
      "step": 6255
    },
    {
      "entropy": 5.310463285446167,
      "epoch": 0.5228867357166722,
      "grad_norm": 1.234375,
      "learning_rate": 0.000497824746821828,
      "loss": 5.2711,
      "mean_token_accuracy": 0.18171454966068268,
      "num_tokens": 11523195.0,
      "step": 6260
    },
    {
      "entropy": 5.38994460105896,
      "epoch": 0.5233043768793852,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004978206153004117,
      "loss": 5.3012,
      "mean_token_accuracy": 0.18391313403844833,
      "num_tokens": 11532348.0,
      "step": 6265
    },
    {
      "entropy": 5.447969818115235,
      "epoch": 0.5237220180420983,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004978164798782523,
      "loss": 5.3111,
      "mean_token_accuracy": 0.18087030500173568,
      "num_tokens": 11542356.0,
      "step": 6270
    },
    {
      "entropy": 5.358577871322632,
      "epoch": 0.5241396592048112,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004978123405554221,
      "loss": 5.316,
      "mean_token_accuracy": 0.18624789118766785,
      "num_tokens": 11550914.0,
      "step": 6275
    },
    {
      "entropy": 5.357284021377564,
      "epoch": 0.5245573003675242,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004978081973319938,
      "loss": 5.2524,
      "mean_token_accuracy": 0.1887720912694931,
      "num_tokens": 11559314.0,
      "step": 6280
    },
    {
      "entropy": 5.298609638214112,
      "epoch": 0.5249749415302373,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004978040502080398,
      "loss": 5.222,
      "mean_token_accuracy": 0.18396988213062287,
      "num_tokens": 11567990.0,
      "step": 6285
    },
    {
      "entropy": 5.325413465499878,
      "epoch": 0.5253925826929502,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004977998991836327,
      "loss": 5.2682,
      "mean_token_accuracy": 0.1804572269320488,
      "num_tokens": 11577133.0,
      "step": 6290
    },
    {
      "entropy": 5.404793977737427,
      "epoch": 0.5258102238556632,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004977957442588451,
      "loss": 5.322,
      "mean_token_accuracy": 0.18051642328500747,
      "num_tokens": 11585330.0,
      "step": 6295
    },
    {
      "entropy": 5.431662559509277,
      "epoch": 0.5262278650183763,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004977915854337498,
      "loss": 5.2855,
      "mean_token_accuracy": 0.17675420343875886,
      "num_tokens": 11595005.0,
      "step": 6300
    },
    {
      "entropy": 5.281835603713989,
      "epoch": 0.5266455061810892,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004977874227084197,
      "loss": 5.1744,
      "mean_token_accuracy": 0.189839705824852,
      "num_tokens": 11603969.0,
      "step": 6305
    },
    {
      "entropy": 5.262091064453125,
      "epoch": 0.5270631473438022,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004977832560829277,
      "loss": 5.2292,
      "mean_token_accuracy": 0.18192449510097503,
      "num_tokens": 11613118.0,
      "step": 6310
    },
    {
      "entropy": 5.3321326732635494,
      "epoch": 0.5274807885065153,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004977790855573464,
      "loss": 5.1735,
      "mean_token_accuracy": 0.1879858985543251,
      "num_tokens": 11621799.0,
      "step": 6315
    },
    {
      "entropy": 5.27553653717041,
      "epoch": 0.5278984296692282,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004977749111317492,
      "loss": 5.2553,
      "mean_token_accuracy": 0.18213354647159577,
      "num_tokens": 11630557.0,
      "step": 6320
    },
    {
      "entropy": 5.365929841995239,
      "epoch": 0.5283160708319412,
      "grad_norm": 1.21875,
      "learning_rate": 0.000497770732806209,
      "loss": 5.284,
      "mean_token_accuracy": 0.18342802226543425,
      "num_tokens": 11639179.0,
      "step": 6325
    },
    {
      "entropy": 5.4187873840332035,
      "epoch": 0.5287337119946542,
      "grad_norm": 1.5859375,
      "learning_rate": 0.000497766550580799,
      "loss": 5.3909,
      "mean_token_accuracy": 0.17303578406572342,
      "num_tokens": 11648273.0,
      "step": 6330
    },
    {
      "entropy": 5.367193937301636,
      "epoch": 0.5291513531573672,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004977623644555923,
      "loss": 5.2713,
      "mean_token_accuracy": 0.18123112320899964,
      "num_tokens": 11656372.0,
      "step": 6335
    },
    {
      "entropy": 5.342953634262085,
      "epoch": 0.5295689943200802,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004977581744306624,
      "loss": 5.2133,
      "mean_token_accuracy": 0.183183516561985,
      "num_tokens": 11665007.0,
      "step": 6340
    },
    {
      "entropy": 5.353275775909424,
      "epoch": 0.5299866354827932,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004977539805060824,
      "loss": 5.3477,
      "mean_token_accuracy": 0.17437319308519364,
      "num_tokens": 11674519.0,
      "step": 6345
    },
    {
      "entropy": 5.465643739700317,
      "epoch": 0.5304042766455062,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004977497826819259,
      "loss": 5.2697,
      "mean_token_accuracy": 0.1804721772670746,
      "num_tokens": 11683705.0,
      "step": 6350
    },
    {
      "entropy": 5.33522515296936,
      "epoch": 0.5308219178082192,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004977455809582664,
      "loss": 5.2339,
      "mean_token_accuracy": 0.18320189267396927,
      "num_tokens": 11693152.0,
      "step": 6355
    },
    {
      "entropy": 5.305587673187256,
      "epoch": 0.5312395589709322,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004977413753351772,
      "loss": 5.2638,
      "mean_token_accuracy": 0.17687471807003022,
      "num_tokens": 11702584.0,
      "step": 6360
    },
    {
      "entropy": 5.334028148651123,
      "epoch": 0.5316572001336451,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004977371658127323,
      "loss": 5.2967,
      "mean_token_accuracy": 0.18502158522605897,
      "num_tokens": 11711633.0,
      "step": 6365
    },
    {
      "entropy": 5.350752305984497,
      "epoch": 0.5320748412963582,
      "grad_norm": 1.125,
      "learning_rate": 0.0004977329523910051,
      "loss": 5.247,
      "mean_token_accuracy": 0.18760232329368592,
      "num_tokens": 11720116.0,
      "step": 6370
    },
    {
      "entropy": 5.427316093444825,
      "epoch": 0.5324924824590712,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004977287350700695,
      "loss": 5.3263,
      "mean_token_accuracy": 0.18175739645957947,
      "num_tokens": 11729564.0,
      "step": 6375
    },
    {
      "entropy": 5.362538194656372,
      "epoch": 0.5329101236217841,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004977245138499993,
      "loss": 5.2129,
      "mean_token_accuracy": 0.19303394705057145,
      "num_tokens": 11737935.0,
      "step": 6380
    },
    {
      "entropy": 5.3124712944030765,
      "epoch": 0.5333277647844972,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004977202887308684,
      "loss": 5.1855,
      "mean_token_accuracy": 0.19043774902820587,
      "num_tokens": 11745989.0,
      "step": 6385
    },
    {
      "entropy": 5.367902755737305,
      "epoch": 0.5337454059472102,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004977160597127507,
      "loss": 5.3527,
      "mean_token_accuracy": 0.18221916258335114,
      "num_tokens": 11755530.0,
      "step": 6390
    },
    {
      "entropy": 5.382074356079102,
      "epoch": 0.5341630471099231,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004977118267957204,
      "loss": 5.32,
      "mean_token_accuracy": 0.17612025141716003,
      "num_tokens": 11764546.0,
      "step": 6395
    },
    {
      "entropy": 5.494977712631226,
      "epoch": 0.5345806882726362,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004977075899798513,
      "loss": 5.4217,
      "mean_token_accuracy": 0.17207112312316894,
      "num_tokens": 11774117.0,
      "step": 6400
    },
    {
      "entropy": 5.417264699935913,
      "epoch": 0.5349983294353492,
      "grad_norm": 1.1875,
      "learning_rate": 0.000497703349265218,
      "loss": 5.2117,
      "mean_token_accuracy": 0.18813925087451935,
      "num_tokens": 11782952.0,
      "step": 6405
    },
    {
      "entropy": 5.376785326004028,
      "epoch": 0.5354159705980621,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004976991046518943,
      "loss": 5.3688,
      "mean_token_accuracy": 0.174733866751194,
      "num_tokens": 11792369.0,
      "step": 6410
    },
    {
      "entropy": 5.322673368453979,
      "epoch": 0.5358336117607752,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004976948561399549,
      "loss": 5.2099,
      "mean_token_accuracy": 0.1901026889681816,
      "num_tokens": 11801365.0,
      "step": 6415
    },
    {
      "entropy": 5.384335041046143,
      "epoch": 0.5362512529234882,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004976906037294738,
      "loss": 5.2894,
      "mean_token_accuracy": 0.18374739438295365,
      "num_tokens": 11809785.0,
      "step": 6420
    },
    {
      "entropy": 5.351803398132324,
      "epoch": 0.5366688940862011,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004976863474205257,
      "loss": 5.1829,
      "mean_token_accuracy": 0.18223468214273453,
      "num_tokens": 11817978.0,
      "step": 6425
    },
    {
      "entropy": 5.388811302185059,
      "epoch": 0.5370865352489141,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004976820872131851,
      "loss": 5.2913,
      "mean_token_accuracy": 0.18446996659040452,
      "num_tokens": 11827796.0,
      "step": 6430
    },
    {
      "entropy": 5.229735851287842,
      "epoch": 0.5375041764116272,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004976778231075266,
      "loss": 5.1501,
      "mean_token_accuracy": 0.18932668417692183,
      "num_tokens": 11836439.0,
      "step": 6435
    },
    {
      "entropy": 5.361645174026489,
      "epoch": 0.5379218175743401,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004976735551036245,
      "loss": 5.3524,
      "mean_token_accuracy": 0.17990088164806367,
      "num_tokens": 11845268.0,
      "step": 6440
    },
    {
      "entropy": 5.4373973369598385,
      "epoch": 0.5383394587370531,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000497669283201554,
      "loss": 5.2788,
      "mean_token_accuracy": 0.180202616751194,
      "num_tokens": 11855185.0,
      "step": 6445
    },
    {
      "entropy": 5.31826958656311,
      "epoch": 0.5387570998997662,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004976650074013895,
      "loss": 5.3065,
      "mean_token_accuracy": 0.1824183687567711,
      "num_tokens": 11865073.0,
      "step": 6450
    },
    {
      "entropy": 5.430147266387939,
      "epoch": 0.5391747410624791,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004976607277032061,
      "loss": 5.2958,
      "mean_token_accuracy": 0.18481045365333557,
      "num_tokens": 11875004.0,
      "step": 6455
    },
    {
      "entropy": 5.382980632781982,
      "epoch": 0.5395923822251921,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004976564441070788,
      "loss": 5.2949,
      "mean_token_accuracy": 0.17465188801288606,
      "num_tokens": 11884905.0,
      "step": 6460
    },
    {
      "entropy": 5.386041259765625,
      "epoch": 0.5400100233879052,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004976521566130823,
      "loss": 5.2839,
      "mean_token_accuracy": 0.17858238518238068,
      "num_tokens": 11894669.0,
      "step": 6465
    },
    {
      "entropy": 5.501698303222656,
      "epoch": 0.5404276645506181,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004976478652212918,
      "loss": 5.462,
      "mean_token_accuracy": 0.16954419314861296,
      "num_tokens": 11904127.0,
      "step": 6470
    },
    {
      "entropy": 5.423941802978516,
      "epoch": 0.5408453057133311,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004976435699317824,
      "loss": 5.2784,
      "mean_token_accuracy": 0.18121567368507385,
      "num_tokens": 11912031.0,
      "step": 6475
    },
    {
      "entropy": 5.309411907196045,
      "epoch": 0.5412629468760441,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004976392707446293,
      "loss": 5.1966,
      "mean_token_accuracy": 0.19244254231452942,
      "num_tokens": 11920961.0,
      "step": 6480
    },
    {
      "entropy": 5.443402051925659,
      "epoch": 0.5416805880387571,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004976349676599078,
      "loss": 5.3587,
      "mean_token_accuracy": 0.1785212427377701,
      "num_tokens": 11930454.0,
      "step": 6485
    },
    {
      "entropy": 5.304493045806884,
      "epoch": 0.5420982292014701,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004976306606776932,
      "loss": 5.1429,
      "mean_token_accuracy": 0.18634013682603837,
      "num_tokens": 11938610.0,
      "step": 6490
    },
    {
      "entropy": 5.323086214065552,
      "epoch": 0.5425158703641831,
      "grad_norm": 1.25,
      "learning_rate": 0.0004976263497980609,
      "loss": 5.3255,
      "mean_token_accuracy": 0.18066132068634033,
      "num_tokens": 11947882.0,
      "step": 6495
    },
    {
      "entropy": 5.381497001647949,
      "epoch": 0.5429335115268961,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004976220350210864,
      "loss": 5.3139,
      "mean_token_accuracy": 0.17742136269807815,
      "num_tokens": 11957552.0,
      "step": 6500
    },
    {
      "entropy": 5.378464460372925,
      "epoch": 0.5433511526896091,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004976177163468451,
      "loss": 5.2361,
      "mean_token_accuracy": 0.18173269033432007,
      "num_tokens": 11967350.0,
      "step": 6505
    },
    {
      "entropy": 5.384993410110473,
      "epoch": 0.5437687938523221,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004976133937754128,
      "loss": 5.2433,
      "mean_token_accuracy": 0.18130826205015182,
      "num_tokens": 11976107.0,
      "step": 6510
    },
    {
      "entropy": 5.35005750656128,
      "epoch": 0.544186435015035,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004976090673068651,
      "loss": 5.2349,
      "mean_token_accuracy": 0.18404854983091354,
      "num_tokens": 11985059.0,
      "step": 6515
    },
    {
      "entropy": 5.305514287948609,
      "epoch": 0.5446040761777481,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004976047369412778,
      "loss": 5.2181,
      "mean_token_accuracy": 0.18331118524074555,
      "num_tokens": 11993453.0,
      "step": 6520
    },
    {
      "entropy": 5.250410461425782,
      "epoch": 0.5450217173404611,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004976004026787266,
      "loss": 5.1863,
      "mean_token_accuracy": 0.1889011651277542,
      "num_tokens": 12002524.0,
      "step": 6525
    },
    {
      "entropy": 5.286372709274292,
      "epoch": 0.545439358503174,
      "grad_norm": 1.375,
      "learning_rate": 0.0004975960645192873,
      "loss": 5.2366,
      "mean_token_accuracy": 0.184749273955822,
      "num_tokens": 12011516.0,
      "step": 6530
    },
    {
      "entropy": 5.4061259746551515,
      "epoch": 0.5458569996658871,
      "grad_norm": 1.125,
      "learning_rate": 0.0004975917224630359,
      "loss": 5.2907,
      "mean_token_accuracy": 0.18020799309015273,
      "num_tokens": 12020058.0,
      "step": 6535
    },
    {
      "entropy": 5.466730833053589,
      "epoch": 0.5462746408286001,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004975873765100487,
      "loss": 5.3838,
      "mean_token_accuracy": 0.17894876897335052,
      "num_tokens": 12029167.0,
      "step": 6540
    },
    {
      "entropy": 5.43693175315857,
      "epoch": 0.546692281991313,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004975830266604013,
      "loss": 5.3256,
      "mean_token_accuracy": 0.17956264615058898,
      "num_tokens": 12038298.0,
      "step": 6545
    },
    {
      "entropy": 5.384665489196777,
      "epoch": 0.5471099231540261,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004975786729141702,
      "loss": 5.2651,
      "mean_token_accuracy": 0.17975437343120576,
      "num_tokens": 12047437.0,
      "step": 6550
    },
    {
      "entropy": 5.362612628936768,
      "epoch": 0.5475275643167391,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004975743152714314,
      "loss": 5.2965,
      "mean_token_accuracy": 0.18631380498409272,
      "num_tokens": 12055821.0,
      "step": 6555
    },
    {
      "entropy": 5.327825403213501,
      "epoch": 0.547945205479452,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004975699537322616,
      "loss": 5.1643,
      "mean_token_accuracy": 0.19042067527770995,
      "num_tokens": 12064378.0,
      "step": 6560
    },
    {
      "entropy": 5.298802900314331,
      "epoch": 0.5483628466421651,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004975655882967366,
      "loss": 5.2241,
      "mean_token_accuracy": 0.18299751430749894,
      "num_tokens": 12073110.0,
      "step": 6565
    },
    {
      "entropy": 5.3863519668579105,
      "epoch": 0.5487804878048781,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004975612189649331,
      "loss": 5.2514,
      "mean_token_accuracy": 0.18545016050338745,
      "num_tokens": 12081224.0,
      "step": 6570
    },
    {
      "entropy": 5.374075174331665,
      "epoch": 0.549198128967591,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004975568457369276,
      "loss": 5.3135,
      "mean_token_accuracy": 0.18435729891061783,
      "num_tokens": 12090279.0,
      "step": 6575
    },
    {
      "entropy": 5.382840347290039,
      "epoch": 0.549615770130304,
      "grad_norm": 1.25,
      "learning_rate": 0.0004975524686127966,
      "loss": 5.3086,
      "mean_token_accuracy": 0.1836516499519348,
      "num_tokens": 12098818.0,
      "step": 6580
    },
    {
      "entropy": 5.343344402313233,
      "epoch": 0.5500334112930171,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004975480875926168,
      "loss": 5.246,
      "mean_token_accuracy": 0.1891148567199707,
      "num_tokens": 12107358.0,
      "step": 6585
    },
    {
      "entropy": 5.371078109741211,
      "epoch": 0.55045105245573,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004975437026764648,
      "loss": 5.3032,
      "mean_token_accuracy": 0.17587835490703582,
      "num_tokens": 12116802.0,
      "step": 6590
    },
    {
      "entropy": 5.356584596633911,
      "epoch": 0.550868693618443,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004975393138644175,
      "loss": 5.2282,
      "mean_token_accuracy": 0.18383945524692535,
      "num_tokens": 12126347.0,
      "step": 6595
    },
    {
      "entropy": 5.264455652236938,
      "epoch": 0.5512863347811561,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004975349211565515,
      "loss": 5.2011,
      "mean_token_accuracy": 0.1821246027946472,
      "num_tokens": 12136138.0,
      "step": 6600
    },
    {
      "entropy": 5.329731607437134,
      "epoch": 0.551703975943869,
      "grad_norm": 1.25,
      "learning_rate": 0.000497530524552944,
      "loss": 5.2707,
      "mean_token_accuracy": 0.18586531579494475,
      "num_tokens": 12146436.0,
      "step": 6605
    },
    {
      "entropy": 5.425878047943115,
      "epoch": 0.552121617106582,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004975261240536716,
      "loss": 5.2802,
      "mean_token_accuracy": 0.18001115024089814,
      "num_tokens": 12157003.0,
      "step": 6610
    },
    {
      "entropy": 5.343220472335815,
      "epoch": 0.5525392582692951,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004975217196588117,
      "loss": 5.2183,
      "mean_token_accuracy": 0.19267099499702453,
      "num_tokens": 12166671.0,
      "step": 6615
    },
    {
      "entropy": 5.276378631591797,
      "epoch": 0.552956899432008,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004975173113684413,
      "loss": 5.2552,
      "mean_token_accuracy": 0.18465711027383805,
      "num_tokens": 12176474.0,
      "step": 6620
    },
    {
      "entropy": 5.390606307983399,
      "epoch": 0.553374540594721,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004975128991826374,
      "loss": 5.2476,
      "mean_token_accuracy": 0.19043864756822587,
      "num_tokens": 12185307.0,
      "step": 6625
    },
    {
      "entropy": 5.351818180084228,
      "epoch": 0.553792181757434,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004975084831014774,
      "loss": 5.2267,
      "mean_token_accuracy": 0.18219854831695556,
      "num_tokens": 12195179.0,
      "step": 6630
    },
    {
      "entropy": 5.296496772766114,
      "epoch": 0.554209822920147,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004975040631250386,
      "loss": 5.2841,
      "mean_token_accuracy": 0.19119415134191514,
      "num_tokens": 12204083.0,
      "step": 6635
    },
    {
      "entropy": 5.29406852722168,
      "epoch": 0.55462746408286,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004974996392533983,
      "loss": 5.2206,
      "mean_token_accuracy": 0.19135015606880187,
      "num_tokens": 12213322.0,
      "step": 6640
    },
    {
      "entropy": 5.323991298675537,
      "epoch": 0.555045105245573,
      "grad_norm": 1.234375,
      "learning_rate": 0.000497495211486634,
      "loss": 5.2148,
      "mean_token_accuracy": 0.18611935675144195,
      "num_tokens": 12222305.0,
      "step": 6645
    },
    {
      "entropy": 5.371159934997559,
      "epoch": 0.555462746408286,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004974907798248232,
      "loss": 5.3233,
      "mean_token_accuracy": 0.17673341184854507,
      "num_tokens": 12232196.0,
      "step": 6650
    },
    {
      "entropy": 5.411021757125854,
      "epoch": 0.555880387570999,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004974863442680434,
      "loss": 5.2521,
      "mean_token_accuracy": 0.18408025801181793,
      "num_tokens": 12241914.0,
      "step": 6655
    },
    {
      "entropy": 5.189435577392578,
      "epoch": 0.556298028733712,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004974819048163725,
      "loss": 5.1464,
      "mean_token_accuracy": 0.1912049800157547,
      "num_tokens": 12250185.0,
      "step": 6660
    },
    {
      "entropy": 5.365625524520874,
      "epoch": 0.556715669896425,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000497477461469888,
      "loss": 5.3231,
      "mean_token_accuracy": 0.17474996000528337,
      "num_tokens": 12258954.0,
      "step": 6665
    },
    {
      "entropy": 5.473125267028808,
      "epoch": 0.557133311059138,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004974730142286677,
      "loss": 5.3323,
      "mean_token_accuracy": 0.17727245539426803,
      "num_tokens": 12268320.0,
      "step": 6670
    },
    {
      "entropy": 5.249721670150757,
      "epoch": 0.557550952221851,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004974685630927895,
      "loss": 5.2127,
      "mean_token_accuracy": 0.1804267629981041,
      "num_tokens": 12276804.0,
      "step": 6675
    },
    {
      "entropy": 5.442297840118409,
      "epoch": 0.557968593384564,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004974641080623314,
      "loss": 5.2641,
      "mean_token_accuracy": 0.1770022466778755,
      "num_tokens": 12286024.0,
      "step": 6680
    },
    {
      "entropy": 5.349287652969361,
      "epoch": 0.558386234547277,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004974596491373711,
      "loss": 5.301,
      "mean_token_accuracy": 0.18299159854650499,
      "num_tokens": 12295154.0,
      "step": 6685
    },
    {
      "entropy": 5.387079477310181,
      "epoch": 0.55880387570999,
      "grad_norm": 1.25,
      "learning_rate": 0.0004974551863179871,
      "loss": 5.363,
      "mean_token_accuracy": 0.17097254991531372,
      "num_tokens": 12306349.0,
      "step": 6690
    },
    {
      "entropy": 5.407424402236939,
      "epoch": 0.5592215168727029,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004974507196042572,
      "loss": 5.3179,
      "mean_token_accuracy": 0.18425286263227464,
      "num_tokens": 12315496.0,
      "step": 6695
    },
    {
      "entropy": 5.2734064102172855,
      "epoch": 0.559639158035416,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004974462489962597,
      "loss": 5.1536,
      "mean_token_accuracy": 0.1862584874033928,
      "num_tokens": 12325070.0,
      "step": 6700
    },
    {
      "entropy": 5.346486568450928,
      "epoch": 0.560056799198129,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004974417744940729,
      "loss": 5.2523,
      "mean_token_accuracy": 0.17760104089975357,
      "num_tokens": 12333509.0,
      "step": 6705
    },
    {
      "entropy": 5.345494794845581,
      "epoch": 0.5604744403608419,
      "grad_norm": 1.140625,
      "learning_rate": 0.000497437296097775,
      "loss": 5.3067,
      "mean_token_accuracy": 0.17850659936666488,
      "num_tokens": 12342495.0,
      "step": 6710
    },
    {
      "entropy": 5.327435064315796,
      "epoch": 0.560892081523555,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004974328138074445,
      "loss": 5.2116,
      "mean_token_accuracy": 0.18249644935131074,
      "num_tokens": 12351560.0,
      "step": 6715
    },
    {
      "entropy": 5.380244398117066,
      "epoch": 0.561309722686268,
      "grad_norm": 1.25,
      "learning_rate": 0.0004974283276231599,
      "loss": 5.3262,
      "mean_token_accuracy": 0.18469928950071335,
      "num_tokens": 12360773.0,
      "step": 6720
    },
    {
      "entropy": 5.357887172698975,
      "epoch": 0.5617273638489809,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004974238375449996,
      "loss": 5.3384,
      "mean_token_accuracy": 0.1726855531334877,
      "num_tokens": 12370519.0,
      "step": 6725
    },
    {
      "entropy": 5.410505056381226,
      "epoch": 0.562145005011694,
      "grad_norm": 1.125,
      "learning_rate": 0.0004974193435730424,
      "loss": 5.3248,
      "mean_token_accuracy": 0.17881592810153962,
      "num_tokens": 12380563.0,
      "step": 6730
    },
    {
      "entropy": 5.455129098892212,
      "epoch": 0.562562646174407,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004974148457073668,
      "loss": 5.3182,
      "mean_token_accuracy": 0.18274307399988174,
      "num_tokens": 12389291.0,
      "step": 6735
    },
    {
      "entropy": 5.329137706756592,
      "epoch": 0.5629802873371199,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004974103439480517,
      "loss": 5.2905,
      "mean_token_accuracy": 0.1844645231962204,
      "num_tokens": 12399037.0,
      "step": 6740
    },
    {
      "entropy": 5.297447586059571,
      "epoch": 0.563397928499833,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004974058382951757,
      "loss": 5.1102,
      "mean_token_accuracy": 0.19304572641849518,
      "num_tokens": 12407553.0,
      "step": 6745
    },
    {
      "entropy": 5.249135494232178,
      "epoch": 0.563815569662546,
      "grad_norm": 1.25,
      "learning_rate": 0.0004974013287488177,
      "loss": 5.1668,
      "mean_token_accuracy": 0.18568285256624223,
      "num_tokens": 12416298.0,
      "step": 6750
    },
    {
      "entropy": 5.363959169387817,
      "epoch": 0.5642332108252589,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004973968153090569,
      "loss": 5.3317,
      "mean_token_accuracy": 0.1737111657857895,
      "num_tokens": 12425918.0,
      "step": 6755
    },
    {
      "entropy": 5.376228904724121,
      "epoch": 0.5646508519879719,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004973922979759722,
      "loss": 5.2656,
      "mean_token_accuracy": 0.17900831997394562,
      "num_tokens": 12434693.0,
      "step": 6760
    },
    {
      "entropy": 5.233968305587768,
      "epoch": 0.565068493150685,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004973877767496424,
      "loss": 5.1272,
      "mean_token_accuracy": 0.1882891222834587,
      "num_tokens": 12444055.0,
      "step": 6765
    },
    {
      "entropy": 5.2691069602966305,
      "epoch": 0.5654861343133979,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004973832516301472,
      "loss": 5.2049,
      "mean_token_accuracy": 0.1932446539402008,
      "num_tokens": 12453012.0,
      "step": 6770
    },
    {
      "entropy": 5.294369649887085,
      "epoch": 0.5659037754761109,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004973787226175652,
      "loss": 5.2364,
      "mean_token_accuracy": 0.18431908637285233,
      "num_tokens": 12462292.0,
      "step": 6775
    },
    {
      "entropy": 5.306050586700439,
      "epoch": 0.566321416638824,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004973741897119761,
      "loss": 5.2026,
      "mean_token_accuracy": 0.19394677579402925,
      "num_tokens": 12471178.0,
      "step": 6780
    },
    {
      "entropy": 5.265408897399903,
      "epoch": 0.5667390578015369,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004973696529134592,
      "loss": 5.2427,
      "mean_token_accuracy": 0.188844895362854,
      "num_tokens": 12479977.0,
      "step": 6785
    },
    {
      "entropy": 5.313236951828003,
      "epoch": 0.5671566989642499,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004973651122220938,
      "loss": 5.296,
      "mean_token_accuracy": 0.17963682264089584,
      "num_tokens": 12489004.0,
      "step": 6790
    },
    {
      "entropy": 5.411258029937744,
      "epoch": 0.567574340126963,
      "grad_norm": 1.25,
      "learning_rate": 0.0004973605676379595,
      "loss": 5.2956,
      "mean_token_accuracy": 0.18012158274650575,
      "num_tokens": 12498979.0,
      "step": 6795
    },
    {
      "entropy": 5.383917236328125,
      "epoch": 0.5679919812896759,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004973560191611357,
      "loss": 5.3565,
      "mean_token_accuracy": 0.17776286751031875,
      "num_tokens": 12508293.0,
      "step": 6800
    },
    {
      "entropy": 5.49052324295044,
      "epoch": 0.5684096224523889,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004973514667917021,
      "loss": 5.3776,
      "mean_token_accuracy": 0.17675091177225113,
      "num_tokens": 12517828.0,
      "step": 6805
    },
    {
      "entropy": 5.370597171783447,
      "epoch": 0.5688272636151019,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004973469105297386,
      "loss": 5.2195,
      "mean_token_accuracy": 0.18173562586307526,
      "num_tokens": 12526569.0,
      "step": 6810
    },
    {
      "entropy": 5.390897464752197,
      "epoch": 0.5692449047778149,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004973423503753247,
      "loss": 5.3547,
      "mean_token_accuracy": 0.18119753003120423,
      "num_tokens": 12535706.0,
      "step": 6815
    },
    {
      "entropy": 5.430558538436889,
      "epoch": 0.5696625459405279,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004973377863285403,
      "loss": 5.244,
      "mean_token_accuracy": 0.18043635040521622,
      "num_tokens": 12544689.0,
      "step": 6820
    },
    {
      "entropy": 5.304398488998413,
      "epoch": 0.5700801871032409,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004973332183894653,
      "loss": 5.2211,
      "mean_token_accuracy": 0.18837016820907593,
      "num_tokens": 12553318.0,
      "step": 6825
    },
    {
      "entropy": 5.292221403121948,
      "epoch": 0.5704978282659539,
      "grad_norm": 1.25,
      "learning_rate": 0.0004973286465581798,
      "loss": 5.1887,
      "mean_token_accuracy": 0.19200338721275328,
      "num_tokens": 12562009.0,
      "step": 6830
    },
    {
      "entropy": 5.388770389556885,
      "epoch": 0.5709154694286669,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004973240708347637,
      "loss": 5.3256,
      "mean_token_accuracy": 0.19126973748207093,
      "num_tokens": 12571043.0,
      "step": 6835
    },
    {
      "entropy": 5.465327453613281,
      "epoch": 0.5713331105913799,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004973194912192971,
      "loss": 5.2619,
      "mean_token_accuracy": 0.18441184908151625,
      "num_tokens": 12581151.0,
      "step": 6840
    },
    {
      "entropy": 5.317709350585938,
      "epoch": 0.5717507517540928,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004973149077118602,
      "loss": 5.2933,
      "mean_token_accuracy": 0.1829424634575844,
      "num_tokens": 12590671.0,
      "step": 6845
    },
    {
      "entropy": 5.315968370437622,
      "epoch": 0.5721683929168059,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004973103203125333,
      "loss": 5.3302,
      "mean_token_accuracy": 0.17839666903018953,
      "num_tokens": 12599693.0,
      "step": 6850
    },
    {
      "entropy": 5.390160179138183,
      "epoch": 0.5725860340795189,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004973057290213965,
      "loss": 5.2248,
      "mean_token_accuracy": 0.1872655600309372,
      "num_tokens": 12609238.0,
      "step": 6855
    },
    {
      "entropy": 5.335610485076904,
      "epoch": 0.5730036752422318,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004973011338385304,
      "loss": 5.2318,
      "mean_token_accuracy": 0.1856747567653656,
      "num_tokens": 12618273.0,
      "step": 6860
    },
    {
      "entropy": 5.321760749816894,
      "epoch": 0.5734213164049449,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004972965347640154,
      "loss": 5.2957,
      "mean_token_accuracy": 0.18263923972845078,
      "num_tokens": 12628229.0,
      "step": 6865
    },
    {
      "entropy": 5.368953704833984,
      "epoch": 0.5738389575676579,
      "grad_norm": 1.234375,
      "learning_rate": 0.000497291931797932,
      "loss": 5.1923,
      "mean_token_accuracy": 0.191641004383564,
      "num_tokens": 12636533.0,
      "step": 6870
    },
    {
      "entropy": 5.273049211502075,
      "epoch": 0.5742565987303708,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004972873249403608,
      "loss": 5.2392,
      "mean_token_accuracy": 0.18528711795806885,
      "num_tokens": 12645503.0,
      "step": 6875
    },
    {
      "entropy": 5.34866943359375,
      "epoch": 0.5746742398930839,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004972827141913822,
      "loss": 5.2341,
      "mean_token_accuracy": 0.17654672712087632,
      "num_tokens": 12654355.0,
      "step": 6880
    },
    {
      "entropy": 5.361291599273682,
      "epoch": 0.5750918810557969,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004972780995510773,
      "loss": 5.2141,
      "mean_token_accuracy": 0.18175720870494844,
      "num_tokens": 12663774.0,
      "step": 6885
    },
    {
      "entropy": 5.1940025806427,
      "epoch": 0.5755095222185098,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004972734810195266,
      "loss": 5.2585,
      "mean_token_accuracy": 0.18426082879304886,
      "num_tokens": 12674436.0,
      "step": 6890
    },
    {
      "entropy": 5.3957408428192135,
      "epoch": 0.5759271633812229,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004972688585968112,
      "loss": 5.1981,
      "mean_token_accuracy": 0.18481385558843613,
      "num_tokens": 12682891.0,
      "step": 6895
    },
    {
      "entropy": 5.255229520797729,
      "epoch": 0.5763448045439359,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004972642322830118,
      "loss": 5.1009,
      "mean_token_accuracy": 0.19705357104539872,
      "num_tokens": 12691170.0,
      "step": 6900
    },
    {
      "entropy": 5.273286390304565,
      "epoch": 0.5767624457066488,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004972596020782096,
      "loss": 5.1984,
      "mean_token_accuracy": 0.18740458488464357,
      "num_tokens": 12700314.0,
      "step": 6905
    },
    {
      "entropy": 5.313592338562012,
      "epoch": 0.5771800868693618,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004972549679824853,
      "loss": 5.2461,
      "mean_token_accuracy": 0.18948954790830613,
      "num_tokens": 12709418.0,
      "step": 6910
    },
    {
      "entropy": 5.333023357391357,
      "epoch": 0.5775977280320749,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004972503299959204,
      "loss": 5.2355,
      "mean_token_accuracy": 0.18070902824401855,
      "num_tokens": 12719287.0,
      "step": 6915
    },
    {
      "entropy": 5.322830009460449,
      "epoch": 0.5780153691947878,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000497245688118596,
      "loss": 5.1621,
      "mean_token_accuracy": 0.19228754341602325,
      "num_tokens": 12728836.0,
      "step": 6920
    },
    {
      "entropy": 5.353468704223633,
      "epoch": 0.5784330103575008,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004972410423505932,
      "loss": 5.293,
      "mean_token_accuracy": 0.17521052658557892,
      "num_tokens": 12738173.0,
      "step": 6925
    },
    {
      "entropy": 5.291993188858032,
      "epoch": 0.5788506515202139,
      "grad_norm": 1.125,
      "learning_rate": 0.0004972363926919935,
      "loss": 5.2725,
      "mean_token_accuracy": 0.1864841639995575,
      "num_tokens": 12747428.0,
      "step": 6930
    },
    {
      "entropy": 5.403028964996338,
      "epoch": 0.5792682926829268,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004972317391428782,
      "loss": 5.3284,
      "mean_token_accuracy": 0.17580013126134872,
      "num_tokens": 12757148.0,
      "step": 6935
    },
    {
      "entropy": 5.454678773880005,
      "epoch": 0.5796859338456398,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004972270817033288,
      "loss": 5.3524,
      "mean_token_accuracy": 0.17018143981695175,
      "num_tokens": 12766820.0,
      "step": 6940
    },
    {
      "entropy": 5.429720783233643,
      "epoch": 0.5801035750083529,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004972224203734269,
      "loss": 5.3264,
      "mean_token_accuracy": 0.17791177928447724,
      "num_tokens": 12776082.0,
      "step": 6945
    },
    {
      "entropy": 5.327076101303101,
      "epoch": 0.5805212161710658,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000497217755153254,
      "loss": 5.299,
      "mean_token_accuracy": 0.18474810421466828,
      "num_tokens": 12785057.0,
      "step": 6950
    },
    {
      "entropy": 5.308185148239136,
      "epoch": 0.5809388573337788,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004972130860428918,
      "loss": 5.2988,
      "mean_token_accuracy": 0.190691839158535,
      "num_tokens": 12793599.0,
      "step": 6955
    },
    {
      "entropy": 5.3690268993377686,
      "epoch": 0.5813564984964918,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004972084130424222,
      "loss": 5.2757,
      "mean_token_accuracy": 0.17726799845695496,
      "num_tokens": 12803775.0,
      "step": 6960
    },
    {
      "entropy": 5.348795747756958,
      "epoch": 0.5817741396592048,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004972037361519267,
      "loss": 5.2193,
      "mean_token_accuracy": 0.1903219923377037,
      "num_tokens": 12812327.0,
      "step": 6965
    },
    {
      "entropy": 5.30856499671936,
      "epoch": 0.5821917808219178,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004971990553714874,
      "loss": 5.1789,
      "mean_token_accuracy": 0.1907658338546753,
      "num_tokens": 12821218.0,
      "step": 6970
    },
    {
      "entropy": 5.371193265914917,
      "epoch": 0.5826094219846308,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004971943707011861,
      "loss": 5.2847,
      "mean_token_accuracy": 0.1793219655752182,
      "num_tokens": 12831579.0,
      "step": 6975
    },
    {
      "entropy": 5.425595474243164,
      "epoch": 0.5830270631473438,
      "grad_norm": 1.25,
      "learning_rate": 0.000497189682141105,
      "loss": 5.3221,
      "mean_token_accuracy": 0.18375834077596664,
      "num_tokens": 12841026.0,
      "step": 6980
    },
    {
      "entropy": 5.297035121917725,
      "epoch": 0.5834447043100568,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004971849896913259,
      "loss": 5.2713,
      "mean_token_accuracy": 0.17574221193790435,
      "num_tokens": 12850136.0,
      "step": 6985
    },
    {
      "entropy": 5.46290545463562,
      "epoch": 0.5838623454727698,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004971802933519312,
      "loss": 5.338,
      "mean_token_accuracy": 0.17922402769327164,
      "num_tokens": 12860345.0,
      "step": 6990
    },
    {
      "entropy": 5.307545185089111,
      "epoch": 0.5842799866354828,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004971755931230031,
      "loss": 5.2674,
      "mean_token_accuracy": 0.18393541276454925,
      "num_tokens": 12868995.0,
      "step": 6995
    },
    {
      "entropy": 5.403428411483764,
      "epoch": 0.5846976277981958,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004971708890046236,
      "loss": 5.415,
      "mean_token_accuracy": 0.17389337867498397,
      "num_tokens": 12878043.0,
      "step": 7000
    },
    {
      "entropy": 5.336083650588989,
      "epoch": 0.5851152689609088,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004971661809968755,
      "loss": 5.2458,
      "mean_token_accuracy": 0.18721241801977156,
      "num_tokens": 12887090.0,
      "step": 7005
    },
    {
      "entropy": 5.278947591781616,
      "epoch": 0.5855329101236217,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004971614690998409,
      "loss": 5.1714,
      "mean_token_accuracy": 0.19780073016881944,
      "num_tokens": 12896349.0,
      "step": 7010
    },
    {
      "entropy": 5.3414857387542725,
      "epoch": 0.5859505512863348,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004971567533136024,
      "loss": 5.2309,
      "mean_token_accuracy": 0.17909024357795716,
      "num_tokens": 12905698.0,
      "step": 7015
    },
    {
      "entropy": 5.253928852081299,
      "epoch": 0.5863681924490478,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004971520336382424,
      "loss": 5.2223,
      "mean_token_accuracy": 0.19666144400835037,
      "num_tokens": 12915146.0,
      "step": 7020
    },
    {
      "entropy": 5.181387281417846,
      "epoch": 0.5867858336117607,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004971473100738437,
      "loss": 5.0843,
      "mean_token_accuracy": 0.1966075822710991,
      "num_tokens": 12924933.0,
      "step": 7025
    },
    {
      "entropy": 5.397075414657593,
      "epoch": 0.5872034747744738,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000497142582620489,
      "loss": 5.331,
      "mean_token_accuracy": 0.17998473197221757,
      "num_tokens": 12934959.0,
      "step": 7030
    },
    {
      "entropy": 5.340605974197388,
      "epoch": 0.5876211159371868,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004971378512782609,
      "loss": 5.1733,
      "mean_token_accuracy": 0.1947772204875946,
      "num_tokens": 12943470.0,
      "step": 7035
    },
    {
      "entropy": 5.223190832138061,
      "epoch": 0.5880387570998997,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004971331160472424,
      "loss": 5.186,
      "mean_token_accuracy": 0.1917681023478508,
      "num_tokens": 12953592.0,
      "step": 7040
    },
    {
      "entropy": 5.3833833694458,
      "epoch": 0.5884563982626128,
      "grad_norm": 1.25,
      "learning_rate": 0.0004971283769275163,
      "loss": 5.3014,
      "mean_token_accuracy": 0.1722431391477585,
      "num_tokens": 12962893.0,
      "step": 7045
    },
    {
      "entropy": 5.354134368896484,
      "epoch": 0.5888740394253258,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004971236339191656,
      "loss": 5.1945,
      "mean_token_accuracy": 0.1897740438580513,
      "num_tokens": 12971635.0,
      "step": 7050
    },
    {
      "entropy": 5.239306449890137,
      "epoch": 0.5892916805880387,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004971188870222734,
      "loss": 5.1704,
      "mean_token_accuracy": 0.1947092667222023,
      "num_tokens": 12981040.0,
      "step": 7055
    },
    {
      "entropy": 5.322380304336548,
      "epoch": 0.5897093217507517,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004971141362369226,
      "loss": 5.176,
      "mean_token_accuracy": 0.185821096599102,
      "num_tokens": 12989898.0,
      "step": 7060
    },
    {
      "entropy": 5.376731729507446,
      "epoch": 0.5901269629134648,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004971093815631965,
      "loss": 5.1823,
      "mean_token_accuracy": 0.1879758670926094,
      "num_tokens": 12998584.0,
      "step": 7065
    },
    {
      "entropy": 5.234632062911987,
      "epoch": 0.5905446040761777,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004971046230011783,
      "loss": 5.2264,
      "mean_token_accuracy": 0.18627288043498993,
      "num_tokens": 13008007.0,
      "step": 7070
    },
    {
      "entropy": 5.357740354537964,
      "epoch": 0.5909622452388907,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004970998605509514,
      "loss": 5.3368,
      "mean_token_accuracy": 0.1804517537355423,
      "num_tokens": 13017034.0,
      "step": 7075
    },
    {
      "entropy": 5.495122909545898,
      "epoch": 0.5913798864016038,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000497095094212599,
      "loss": 5.4182,
      "mean_token_accuracy": 0.16964825838804246,
      "num_tokens": 13025916.0,
      "step": 7080
    },
    {
      "entropy": 5.4193336963653564,
      "epoch": 0.5917975275643167,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004970903239862047,
      "loss": 5.2954,
      "mean_token_accuracy": 0.17611705362796784,
      "num_tokens": 13035242.0,
      "step": 7085
    },
    {
      "entropy": 5.313186311721802,
      "epoch": 0.5922151687270297,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004970855498718519,
      "loss": 5.2179,
      "mean_token_accuracy": 0.18447796553373336,
      "num_tokens": 13044123.0,
      "step": 7090
    },
    {
      "entropy": 5.325734806060791,
      "epoch": 0.5926328098897428,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004970807718696243,
      "loss": 5.2604,
      "mean_token_accuracy": 0.1772498980164528,
      "num_tokens": 13053503.0,
      "step": 7095
    },
    {
      "entropy": 5.349041843414307,
      "epoch": 0.5930504510524557,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004970759899796055,
      "loss": 5.2482,
      "mean_token_accuracy": 0.18219223469495774,
      "num_tokens": 13062106.0,
      "step": 7100
    },
    {
      "entropy": 5.385118722915649,
      "epoch": 0.5934680922151687,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004970712042018791,
      "loss": 5.2374,
      "mean_token_accuracy": 0.18351289480924607,
      "num_tokens": 13071509.0,
      "step": 7105
    },
    {
      "entropy": 5.264652824401855,
      "epoch": 0.5938857333778818,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004970664145365289,
      "loss": 5.1821,
      "mean_token_accuracy": 0.19266004711389542,
      "num_tokens": 13080469.0,
      "step": 7110
    },
    {
      "entropy": 5.270586633682251,
      "epoch": 0.5943033745405947,
      "grad_norm": 1.1875,
      "learning_rate": 0.000497061620983639,
      "loss": 5.1381,
      "mean_token_accuracy": 0.19512879103422165,
      "num_tokens": 13089826.0,
      "step": 7115
    },
    {
      "entropy": 5.374065494537353,
      "epoch": 0.5947210157033077,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004970568235432931,
      "loss": 5.3253,
      "mean_token_accuracy": 0.18222101777791977,
      "num_tokens": 13098630.0,
      "step": 7120
    },
    {
      "entropy": 5.358995199203491,
      "epoch": 0.5951386568660207,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004970520222155751,
      "loss": 5.2257,
      "mean_token_accuracy": 0.1871304079890251,
      "num_tokens": 13106819.0,
      "step": 7125
    },
    {
      "entropy": 5.369958782196045,
      "epoch": 0.5955562980287337,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004970472170005692,
      "loss": 5.2379,
      "mean_token_accuracy": 0.18896643668413163,
      "num_tokens": 13115471.0,
      "step": 7130
    },
    {
      "entropy": 5.246409225463867,
      "epoch": 0.5959739391914467,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004970424078983596,
      "loss": 5.1224,
      "mean_token_accuracy": 0.19780913591384888,
      "num_tokens": 13124537.0,
      "step": 7135
    },
    {
      "entropy": 5.279074430465698,
      "epoch": 0.5963915803541597,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004970375949090302,
      "loss": 5.2574,
      "mean_token_accuracy": 0.18253438025712967,
      "num_tokens": 13135326.0,
      "step": 7140
    },
    {
      "entropy": 5.476173019409179,
      "epoch": 0.5968092215168727,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004970327780326655,
      "loss": 5.3551,
      "mean_token_accuracy": 0.17697446942329406,
      "num_tokens": 13145502.0,
      "step": 7145
    },
    {
      "entropy": 5.411634635925293,
      "epoch": 0.5972268626795857,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004970279572693498,
      "loss": 5.258,
      "mean_token_accuracy": 0.18461368680000306,
      "num_tokens": 13154808.0,
      "step": 7150
    },
    {
      "entropy": 5.253088760375976,
      "epoch": 0.5976445038422987,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004970231326191674,
      "loss": 5.3812,
      "mean_token_accuracy": 0.17350238263607026,
      "num_tokens": 13163982.0,
      "step": 7155
    },
    {
      "entropy": 5.361168766021729,
      "epoch": 0.5980621450050116,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004970183040822029,
      "loss": 5.2722,
      "mean_token_accuracy": 0.18401766270399095,
      "num_tokens": 13172446.0,
      "step": 7160
    },
    {
      "entropy": 5.389185571670533,
      "epoch": 0.5984797861677247,
      "grad_norm": 1.125,
      "learning_rate": 0.0004970134716585407,
      "loss": 5.2269,
      "mean_token_accuracy": 0.18320017158985138,
      "num_tokens": 13181671.0,
      "step": 7165
    },
    {
      "entropy": 5.250359344482422,
      "epoch": 0.5988974273304377,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004970086353482654,
      "loss": 5.1613,
      "mean_token_accuracy": 0.1845400884747505,
      "num_tokens": 13190207.0,
      "step": 7170
    },
    {
      "entropy": 5.2645087242126465,
      "epoch": 0.5993150684931506,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004970037951514617,
      "loss": 5.1487,
      "mean_token_accuracy": 0.1919180080294609,
      "num_tokens": 13198866.0,
      "step": 7175
    },
    {
      "entropy": 5.2738477230072025,
      "epoch": 0.5997327096558637,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004969989510682143,
      "loss": 5.187,
      "mean_token_accuracy": 0.19386524707078934,
      "num_tokens": 13207921.0,
      "step": 7180
    },
    {
      "entropy": 5.441553592681885,
      "epoch": 0.6001503508185767,
      "grad_norm": 1.375,
      "learning_rate": 0.0004969941030986082,
      "loss": 5.3503,
      "mean_token_accuracy": 0.1791888102889061,
      "num_tokens": 13217759.0,
      "step": 7185
    },
    {
      "entropy": 5.2828004360198975,
      "epoch": 0.6005679919812896,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000496989251242728,
      "loss": 5.1333,
      "mean_token_accuracy": 0.18978789299726487,
      "num_tokens": 13226592.0,
      "step": 7190
    },
    {
      "entropy": 5.2834490776062015,
      "epoch": 0.6009856331440027,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004969843955006588,
      "loss": 5.1953,
      "mean_token_accuracy": 0.18763985782861708,
      "num_tokens": 13236123.0,
      "step": 7195
    },
    {
      "entropy": 5.330254459381104,
      "epoch": 0.6014032743067157,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004969795358724855,
      "loss": 5.3106,
      "mean_token_accuracy": 0.1803394615650177,
      "num_tokens": 13246169.0,
      "step": 7200
    },
    {
      "entropy": 5.27890911102295,
      "epoch": 0.6018209154694286,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004969746723582932,
      "loss": 5.1732,
      "mean_token_accuracy": 0.19286707192659377,
      "num_tokens": 13256141.0,
      "step": 7205
    },
    {
      "entropy": 5.371651697158813,
      "epoch": 0.6022385566321417,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004969698049581672,
      "loss": 5.3484,
      "mean_token_accuracy": 0.18351904153823853,
      "num_tokens": 13265491.0,
      "step": 7210
    },
    {
      "entropy": 5.493934679031372,
      "epoch": 0.6026561977948547,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004969649336721924,
      "loss": 5.3529,
      "mean_token_accuracy": 0.18758141547441481,
      "num_tokens": 13274805.0,
      "step": 7215
    },
    {
      "entropy": 5.403158807754517,
      "epoch": 0.6030738389575676,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004969600585004544,
      "loss": 5.2959,
      "mean_token_accuracy": 0.1817632332444191,
      "num_tokens": 13284589.0,
      "step": 7220
    },
    {
      "entropy": 5.273687744140625,
      "epoch": 0.6034914801202806,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004969551794430383,
      "loss": 5.1332,
      "mean_token_accuracy": 0.19399268180131912,
      "num_tokens": 13293370.0,
      "step": 7225
    },
    {
      "entropy": 5.3851055145263675,
      "epoch": 0.6039091212829937,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004969502965000297,
      "loss": 5.3173,
      "mean_token_accuracy": 0.18494612872600555,
      "num_tokens": 13302555.0,
      "step": 7230
    },
    {
      "entropy": 5.322985553741455,
      "epoch": 0.6043267624457066,
      "grad_norm": 1.4609375,
      "learning_rate": 0.000496945409671514,
      "loss": 5.235,
      "mean_token_accuracy": 0.1859142526984215,
      "num_tokens": 13311632.0,
      "step": 7235
    },
    {
      "entropy": 5.3157310962677,
      "epoch": 0.6047444036084196,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004969405189575768,
      "loss": 5.1704,
      "mean_token_accuracy": 0.19575650691986085,
      "num_tokens": 13320593.0,
      "step": 7240
    },
    {
      "entropy": 5.257418537139893,
      "epoch": 0.6051620447711327,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004969356243583036,
      "loss": 5.1494,
      "mean_token_accuracy": 0.19478675723075867,
      "num_tokens": 13329095.0,
      "step": 7245
    },
    {
      "entropy": 5.25714225769043,
      "epoch": 0.6055796859338456,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004969307258737802,
      "loss": 5.2143,
      "mean_token_accuracy": 0.189331291615963,
      "num_tokens": 13338404.0,
      "step": 7250
    },
    {
      "entropy": 5.317760467529297,
      "epoch": 0.6059973270965586,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004969258235040921,
      "loss": 5.1934,
      "mean_token_accuracy": 0.1896978050470352,
      "num_tokens": 13347095.0,
      "step": 7255
    },
    {
      "entropy": 5.342493391036987,
      "epoch": 0.6064149682592717,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004969209172493255,
      "loss": 5.2386,
      "mean_token_accuracy": 0.18658900260925293,
      "num_tokens": 13356606.0,
      "step": 7260
    },
    {
      "entropy": 5.346424150466919,
      "epoch": 0.6068326094219846,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004969160071095661,
      "loss": 5.2473,
      "mean_token_accuracy": 0.18290272057056428,
      "num_tokens": 13366889.0,
      "step": 7265
    },
    {
      "entropy": 5.351289844512939,
      "epoch": 0.6072502505846976,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004969110930848999,
      "loss": 5.257,
      "mean_token_accuracy": 0.18321869969367982,
      "num_tokens": 13376761.0,
      "step": 7270
    },
    {
      "entropy": 5.319521951675415,
      "epoch": 0.6076678917474106,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004969061751754129,
      "loss": 5.1648,
      "mean_token_accuracy": 0.19837014973163605,
      "num_tokens": 13386226.0,
      "step": 7275
    },
    {
      "entropy": 5.209369802474976,
      "epoch": 0.6080855329101236,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004969012533811911,
      "loss": 5.1169,
      "mean_token_accuracy": 0.18966408222913742,
      "num_tokens": 13394243.0,
      "step": 7280
    },
    {
      "entropy": 5.3636657238006595,
      "epoch": 0.6085031740728366,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004968963277023208,
      "loss": 5.3816,
      "mean_token_accuracy": 0.17408158481121064,
      "num_tokens": 13403708.0,
      "step": 7285
    },
    {
      "entropy": 5.365334177017212,
      "epoch": 0.6089208152355496,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004968913981388882,
      "loss": 5.225,
      "mean_token_accuracy": 0.18925479352474212,
      "num_tokens": 13412911.0,
      "step": 7290
    },
    {
      "entropy": 5.366679859161377,
      "epoch": 0.6093384563982626,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004968864646909797,
      "loss": 5.3637,
      "mean_token_accuracy": 0.17972162812948228,
      "num_tokens": 13422423.0,
      "step": 7295
    },
    {
      "entropy": 5.320084190368652,
      "epoch": 0.6097560975609756,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004968815273586814,
      "loss": 5.2029,
      "mean_token_accuracy": 0.18353873044252395,
      "num_tokens": 13430984.0,
      "step": 7300
    },
    {
      "entropy": 5.337083768844605,
      "epoch": 0.6101737387236886,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004968765861420799,
      "loss": 5.2573,
      "mean_token_accuracy": 0.1862660065293312,
      "num_tokens": 13440492.0,
      "step": 7305
    },
    {
      "entropy": 5.411310434341431,
      "epoch": 0.6105913798864016,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004968716410412617,
      "loss": 5.2409,
      "mean_token_accuracy": 0.18004440367221833,
      "num_tokens": 13449222.0,
      "step": 7310
    },
    {
      "entropy": 5.309297180175781,
      "epoch": 0.6110090210491146,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004968666920563134,
      "loss": 5.241,
      "mean_token_accuracy": 0.18749442845582961,
      "num_tokens": 13459126.0,
      "step": 7315
    },
    {
      "entropy": 5.402968645095825,
      "epoch": 0.6114266622118276,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004968617391873216,
      "loss": 5.3598,
      "mean_token_accuracy": 0.17306229770183562,
      "num_tokens": 13468811.0,
      "step": 7320
    },
    {
      "entropy": 5.361046934127808,
      "epoch": 0.6118443033745405,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004968567824343729,
      "loss": 5.161,
      "mean_token_accuracy": 0.18235436379909514,
      "num_tokens": 13477765.0,
      "step": 7325
    },
    {
      "entropy": 5.34081015586853,
      "epoch": 0.6122619445372536,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004968518217975542,
      "loss": 5.2991,
      "mean_token_accuracy": 0.1704414039850235,
      "num_tokens": 13487184.0,
      "step": 7330
    },
    {
      "entropy": 5.357620620727539,
      "epoch": 0.6126795856999666,
      "grad_norm": 1.125,
      "learning_rate": 0.0004968468572769524,
      "loss": 5.2466,
      "mean_token_accuracy": 0.18385557383298873,
      "num_tokens": 13496784.0,
      "step": 7335
    },
    {
      "entropy": 5.370666408538819,
      "epoch": 0.6130972268626795,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004968418888726543,
      "loss": 5.2778,
      "mean_token_accuracy": 0.18444303125143052,
      "num_tokens": 13505824.0,
      "step": 7340
    },
    {
      "entropy": 5.3123565196990965,
      "epoch": 0.6135148680253926,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004968369165847469,
      "loss": 5.1317,
      "mean_token_accuracy": 0.18691591322422027,
      "num_tokens": 13515111.0,
      "step": 7345
    },
    {
      "entropy": 5.236346054077148,
      "epoch": 0.6139325091881056,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004968319404133173,
      "loss": 5.1544,
      "mean_token_accuracy": 0.19265615344047546,
      "num_tokens": 13524163.0,
      "step": 7350
    },
    {
      "entropy": 5.241273498535156,
      "epoch": 0.6143501503508185,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004968269603584525,
      "loss": 5.1312,
      "mean_token_accuracy": 0.19498640447854995,
      "num_tokens": 13533368.0,
      "step": 7355
    },
    {
      "entropy": 5.320401906967163,
      "epoch": 0.6147677915135316,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004968219764202397,
      "loss": 5.3421,
      "mean_token_accuracy": 0.1785755604505539,
      "num_tokens": 13543708.0,
      "step": 7360
    },
    {
      "entropy": 5.367575216293335,
      "epoch": 0.6151854326762446,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004968169885987663,
      "loss": 5.2676,
      "mean_token_accuracy": 0.18267862796783446,
      "num_tokens": 13552561.0,
      "step": 7365
    },
    {
      "entropy": 5.4310976505279545,
      "epoch": 0.6156030738389575,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004968119968941193,
      "loss": 5.3533,
      "mean_token_accuracy": 0.17482028156518936,
      "num_tokens": 13562050.0,
      "step": 7370
    },
    {
      "entropy": 5.273569202423095,
      "epoch": 0.6160207150016705,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004968070013063866,
      "loss": 5.1877,
      "mean_token_accuracy": 0.18400951027870177,
      "num_tokens": 13571356.0,
      "step": 7375
    },
    {
      "entropy": 5.262342119216919,
      "epoch": 0.6164383561643836,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004968020018356552,
      "loss": 5.1844,
      "mean_token_accuracy": 0.1795240670442581,
      "num_tokens": 13581357.0,
      "step": 7380
    },
    {
      "entropy": 5.3235832214355465,
      "epoch": 0.6168559973270965,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004967969984820128,
      "loss": 5.1665,
      "mean_token_accuracy": 0.18856770545244217,
      "num_tokens": 13589392.0,
      "step": 7385
    },
    {
      "entropy": 5.297839641571045,
      "epoch": 0.6172736384898095,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004967919912455469,
      "loss": 5.137,
      "mean_token_accuracy": 0.20096796303987502,
      "num_tokens": 13598376.0,
      "step": 7390
    },
    {
      "entropy": 5.174291801452637,
      "epoch": 0.6176912796525226,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004967869801263454,
      "loss": 5.0656,
      "mean_token_accuracy": 0.1944248631596565,
      "num_tokens": 13607014.0,
      "step": 7395
    },
    {
      "entropy": 5.339598274230957,
      "epoch": 0.6181089208152355,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004967819651244957,
      "loss": 5.2361,
      "mean_token_accuracy": 0.18238510191440582,
      "num_tokens": 13615700.0,
      "step": 7400
    },
    {
      "entropy": 5.38496675491333,
      "epoch": 0.6185265619779485,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004967769462400857,
      "loss": 5.2813,
      "mean_token_accuracy": 0.1903863787651062,
      "num_tokens": 13625230.0,
      "step": 7405
    },
    {
      "entropy": 5.4405985355377195,
      "epoch": 0.6189442031406616,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004967719234732033,
      "loss": 5.4258,
      "mean_token_accuracy": 0.1819121539592743,
      "num_tokens": 13635258.0,
      "step": 7410
    },
    {
      "entropy": 5.270614814758301,
      "epoch": 0.6193618443033745,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004967668968239365,
      "loss": 5.1317,
      "mean_token_accuracy": 0.1939717173576355,
      "num_tokens": 13643688.0,
      "step": 7415
    },
    {
      "entropy": 5.303300952911377,
      "epoch": 0.6197794854660875,
      "grad_norm": 1.25,
      "learning_rate": 0.0004967618662923732,
      "loss": 5.2873,
      "mean_token_accuracy": 0.18608999848365784,
      "num_tokens": 13652894.0,
      "step": 7420
    },
    {
      "entropy": 5.313575887680054,
      "epoch": 0.6201971266288006,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004967568318786016,
      "loss": 5.2076,
      "mean_token_accuracy": 0.18977562785148622,
      "num_tokens": 13662852.0,
      "step": 7425
    },
    {
      "entropy": 5.340383863449096,
      "epoch": 0.6206147677915135,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004967517935827097,
      "loss": 5.1882,
      "mean_token_accuracy": 0.18542782515287398,
      "num_tokens": 13671457.0,
      "step": 7430
    },
    {
      "entropy": 5.351501989364624,
      "epoch": 0.6210324089542265,
      "grad_norm": 1.25,
      "learning_rate": 0.0004967467514047856,
      "loss": 5.276,
      "mean_token_accuracy": 0.18492195904254913,
      "num_tokens": 13681002.0,
      "step": 7435
    },
    {
      "entropy": 5.307955884933472,
      "epoch": 0.6214500501169395,
      "grad_norm": 1.25,
      "learning_rate": 0.0004967417053449178,
      "loss": 5.2049,
      "mean_token_accuracy": 0.18511720299720763,
      "num_tokens": 13690677.0,
      "step": 7440
    },
    {
      "entropy": 5.227345323562622,
      "epoch": 0.6218676912796525,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004967366554031945,
      "loss": 5.0412,
      "mean_token_accuracy": 0.19810611456632615,
      "num_tokens": 13699151.0,
      "step": 7445
    },
    {
      "entropy": 5.260391616821289,
      "epoch": 0.6222853324423655,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004967316015797042,
      "loss": 5.2514,
      "mean_token_accuracy": 0.18765032142400742,
      "num_tokens": 13708105.0,
      "step": 7450
    },
    {
      "entropy": 5.3460986614227295,
      "epoch": 0.6227029736050785,
      "grad_norm": 1.25,
      "learning_rate": 0.0004967265438745352,
      "loss": 5.3876,
      "mean_token_accuracy": 0.17505446076393127,
      "num_tokens": 13717094.0,
      "step": 7455
    },
    {
      "entropy": 5.437796640396118,
      "epoch": 0.6231206147677915,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004967214822877763,
      "loss": 5.2305,
      "mean_token_accuracy": 0.19002241641283035,
      "num_tokens": 13726639.0,
      "step": 7460
    },
    {
      "entropy": 5.326229476928711,
      "epoch": 0.6235382559305045,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004967164168195159,
      "loss": 5.1912,
      "mean_token_accuracy": 0.18206896185874938,
      "num_tokens": 13735220.0,
      "step": 7465
    },
    {
      "entropy": 5.171748781204224,
      "epoch": 0.6239558970932175,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004967113474698428,
      "loss": 5.1524,
      "mean_token_accuracy": 0.19392013400793076,
      "num_tokens": 13743277.0,
      "step": 7470
    },
    {
      "entropy": 5.319364356994629,
      "epoch": 0.6243735382559304,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004967062742388456,
      "loss": 5.2173,
      "mean_token_accuracy": 0.1907710313796997,
      "num_tokens": 13752946.0,
      "step": 7475
    },
    {
      "entropy": 5.373605060577392,
      "epoch": 0.6247911794186435,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004967011971266132,
      "loss": 5.2665,
      "mean_token_accuracy": 0.1872479036450386,
      "num_tokens": 13761297.0,
      "step": 7480
    },
    {
      "entropy": 5.347911024093628,
      "epoch": 0.6252088205813565,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004966961161332346,
      "loss": 5.21,
      "mean_token_accuracy": 0.18225371241569518,
      "num_tokens": 13771238.0,
      "step": 7485
    },
    {
      "entropy": 5.307698106765747,
      "epoch": 0.6256264617440696,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004966910312587985,
      "loss": 5.2163,
      "mean_token_accuracy": 0.18507746905088424,
      "num_tokens": 13781420.0,
      "step": 7490
    },
    {
      "entropy": 5.273213529586792,
      "epoch": 0.6260441029067825,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004966859425033941,
      "loss": 5.1881,
      "mean_token_accuracy": 0.19007139801979064,
      "num_tokens": 13789833.0,
      "step": 7495
    },
    {
      "entropy": 5.381387567520141,
      "epoch": 0.6264617440694955,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004966808498671104,
      "loss": 5.3131,
      "mean_token_accuracy": 0.17912554889917373,
      "num_tokens": 13799202.0,
      "step": 7500
    },
    {
      "entropy": 5.402695369720459,
      "epoch": 0.6268793852322085,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004966757533500367,
      "loss": 5.296,
      "mean_token_accuracy": 0.1839716613292694,
      "num_tokens": 13809097.0,
      "step": 7505
    },
    {
      "entropy": 5.3216663837432865,
      "epoch": 0.6272970263949215,
      "grad_norm": 1.203125,
      "learning_rate": 0.000496670652952262,
      "loss": 5.205,
      "mean_token_accuracy": 0.19146456569433212,
      "num_tokens": 13818437.0,
      "step": 7510
    },
    {
      "entropy": 5.237986040115357,
      "epoch": 0.6277146675576345,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004966655486738757,
      "loss": 5.2038,
      "mean_token_accuracy": 0.18793047070503235,
      "num_tokens": 13827083.0,
      "step": 7515
    },
    {
      "entropy": 5.285517644882202,
      "epoch": 0.6281323087203475,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004966604405149672,
      "loss": 5.1805,
      "mean_token_accuracy": 0.18774096220731734,
      "num_tokens": 13835905.0,
      "step": 7520
    },
    {
      "entropy": 5.285591888427734,
      "epoch": 0.6285499498830605,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004966553284756257,
      "loss": 5.1049,
      "mean_token_accuracy": 0.1971011847257614,
      "num_tokens": 13845136.0,
      "step": 7525
    },
    {
      "entropy": 5.212302923202515,
      "epoch": 0.6289675910457735,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004966502125559411,
      "loss": 5.1915,
      "mean_token_accuracy": 0.1868421420454979,
      "num_tokens": 13855087.0,
      "step": 7530
    },
    {
      "entropy": 5.323467683792114,
      "epoch": 0.6293852322084865,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004966450927560026,
      "loss": 5.2415,
      "mean_token_accuracy": 0.18338870257139206,
      "num_tokens": 13864077.0,
      "step": 7535
    },
    {
      "entropy": 5.360993719100952,
      "epoch": 0.6298028733711994,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004966399690759,
      "loss": 5.2401,
      "mean_token_accuracy": 0.18297147303819655,
      "num_tokens": 13873404.0,
      "step": 7540
    },
    {
      "entropy": 5.39419584274292,
      "epoch": 0.6302205145339125,
      "grad_norm": 1.21875,
      "learning_rate": 0.000496634841515723,
      "loss": 5.3104,
      "mean_token_accuracy": 0.1822591856122017,
      "num_tokens": 13883231.0,
      "step": 7545
    },
    {
      "entropy": 5.239055204391479,
      "epoch": 0.6306381556966255,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004966297100755613,
      "loss": 5.0497,
      "mean_token_accuracy": 0.19852859377861024,
      "num_tokens": 13892314.0,
      "step": 7550
    },
    {
      "entropy": 5.205962800979615,
      "epoch": 0.6310557968593384,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004966245747555047,
      "loss": 5.1698,
      "mean_token_accuracy": 0.19216402918100356,
      "num_tokens": 13900771.0,
      "step": 7555
    },
    {
      "entropy": 5.215718221664429,
      "epoch": 0.6314734380220515,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004966194355556432,
      "loss": 5.146,
      "mean_token_accuracy": 0.19043028354644775,
      "num_tokens": 13909615.0,
      "step": 7560
    },
    {
      "entropy": 5.348692321777344,
      "epoch": 0.6318910791847645,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004966142924760666,
      "loss": 5.262,
      "mean_token_accuracy": 0.1833634778857231,
      "num_tokens": 13918491.0,
      "step": 7565
    },
    {
      "entropy": 5.287823867797852,
      "epoch": 0.6323087203474774,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004966091455168652,
      "loss": 5.2582,
      "mean_token_accuracy": 0.17764877378940583,
      "num_tokens": 13927935.0,
      "step": 7570
    },
    {
      "entropy": 5.289534282684326,
      "epoch": 0.6327263615101905,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004966039946781289,
      "loss": 5.0838,
      "mean_token_accuracy": 0.19476981312036515,
      "num_tokens": 13936524.0,
      "step": 7575
    },
    {
      "entropy": 5.252709674835205,
      "epoch": 0.6331440026729035,
      "grad_norm": 1.203125,
      "learning_rate": 0.000496598839959948,
      "loss": 5.1655,
      "mean_token_accuracy": 0.1889199823141098,
      "num_tokens": 13945544.0,
      "step": 7580
    },
    {
      "entropy": 5.272776508331299,
      "epoch": 0.6335616438356164,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004965936813624126,
      "loss": 5.1952,
      "mean_token_accuracy": 0.19397238343954087,
      "num_tokens": 13955016.0,
      "step": 7585
    },
    {
      "entropy": 5.2908196449279785,
      "epoch": 0.6339792849983295,
      "grad_norm": 1.234375,
      "learning_rate": 0.000496588518885613,
      "loss": 5.2104,
      "mean_token_accuracy": 0.18109929263591767,
      "num_tokens": 13964045.0,
      "step": 7590
    },
    {
      "entropy": 5.315970849990845,
      "epoch": 0.6343969261610425,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004965833525296397,
      "loss": 5.21,
      "mean_token_accuracy": 0.18605217188596726,
      "num_tokens": 13973554.0,
      "step": 7595
    },
    {
      "entropy": 5.305334138870239,
      "epoch": 0.6348145673237554,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004965781822945831,
      "loss": 5.2013,
      "mean_token_accuracy": 0.18150089532136918,
      "num_tokens": 13982669.0,
      "step": 7600
    },
    {
      "entropy": 5.348069143295288,
      "epoch": 0.6352322084864684,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004965730081805338,
      "loss": 5.3013,
      "mean_token_accuracy": 0.17758703529834746,
      "num_tokens": 13993392.0,
      "step": 7605
    },
    {
      "entropy": 5.437088966369629,
      "epoch": 0.6356498496491815,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004965678301875821,
      "loss": 5.3089,
      "mean_token_accuracy": 0.1790178969502449,
      "num_tokens": 14002659.0,
      "step": 7610
    },
    {
      "entropy": 5.297514152526856,
      "epoch": 0.6360674908118944,
      "grad_norm": 1.21875,
      "learning_rate": 0.000496562648315819,
      "loss": 5.2413,
      "mean_token_accuracy": 0.18059322685003282,
      "num_tokens": 14011561.0,
      "step": 7615
    },
    {
      "entropy": 5.316247034072876,
      "epoch": 0.6364851319746074,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000496557462565335,
      "loss": 5.2197,
      "mean_token_accuracy": 0.18654428869485856,
      "num_tokens": 14021286.0,
      "step": 7620
    },
    {
      "entropy": 5.288998317718506,
      "epoch": 0.6369027731373205,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004965522729362209,
      "loss": 5.2223,
      "mean_token_accuracy": 0.18602666407823562,
      "num_tokens": 14031901.0,
      "step": 7625
    },
    {
      "entropy": 5.378972339630127,
      "epoch": 0.6373204143000334,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004965470794285676,
      "loss": 5.2814,
      "mean_token_accuracy": 0.18820435404777527,
      "num_tokens": 14041000.0,
      "step": 7630
    },
    {
      "entropy": 5.401868629455566,
      "epoch": 0.6377380554627464,
      "grad_norm": 1.265625,
      "learning_rate": 0.000496541882042466,
      "loss": 5.3044,
      "mean_token_accuracy": 0.17929129302501678,
      "num_tokens": 14050567.0,
      "step": 7635
    },
    {
      "entropy": 5.250955963134766,
      "epoch": 0.6381556966254595,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004965366807780071,
      "loss": 5.1364,
      "mean_token_accuracy": 0.19183983653783798,
      "num_tokens": 14059990.0,
      "step": 7640
    },
    {
      "entropy": 5.326308012008667,
      "epoch": 0.6385733377881724,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004965314756352818,
      "loss": 5.2808,
      "mean_token_accuracy": 0.1800059974193573,
      "num_tokens": 14069317.0,
      "step": 7645
    },
    {
      "entropy": 5.3463438034057615,
      "epoch": 0.6389909789508854,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004965262666143816,
      "loss": 5.2253,
      "mean_token_accuracy": 0.1908731773495674,
      "num_tokens": 14078809.0,
      "step": 7650
    },
    {
      "entropy": 5.339958190917969,
      "epoch": 0.6394086201135984,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004965210537153973,
      "loss": 5.2693,
      "mean_token_accuracy": 0.1820810928940773,
      "num_tokens": 14087660.0,
      "step": 7655
    },
    {
      "entropy": 5.369264554977417,
      "epoch": 0.6398262612763114,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004965158369384204,
      "loss": 5.2631,
      "mean_token_accuracy": 0.1803284004330635,
      "num_tokens": 14096585.0,
      "step": 7660
    },
    {
      "entropy": 5.273446750640869,
      "epoch": 0.6402439024390244,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000496510616283542,
      "loss": 5.1644,
      "mean_token_accuracy": 0.19130808115005493,
      "num_tokens": 14106079.0,
      "step": 7665
    },
    {
      "entropy": 5.353111505508423,
      "epoch": 0.6406615436017374,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004965053917508537,
      "loss": 5.28,
      "mean_token_accuracy": 0.1832178458571434,
      "num_tokens": 14115988.0,
      "step": 7670
    },
    {
      "entropy": 5.34995813369751,
      "epoch": 0.6410791847644504,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004965001633404469,
      "loss": 5.3483,
      "mean_token_accuracy": 0.17697596102952956,
      "num_tokens": 14126980.0,
      "step": 7675
    },
    {
      "entropy": 5.3710198402404785,
      "epoch": 0.6414968259271634,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004964949310524131,
      "loss": 5.2604,
      "mean_token_accuracy": 0.18837666362524033,
      "num_tokens": 14135780.0,
      "step": 7680
    },
    {
      "entropy": 5.394362115859986,
      "epoch": 0.6419144670898764,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004964896948868439,
      "loss": 5.2922,
      "mean_token_accuracy": 0.17141007632017136,
      "num_tokens": 14145655.0,
      "step": 7685
    },
    {
      "entropy": 5.345735120773315,
      "epoch": 0.6423321082525894,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004964844548438311,
      "loss": 5.1735,
      "mean_token_accuracy": 0.18798206001520157,
      "num_tokens": 14154157.0,
      "step": 7690
    },
    {
      "entropy": 5.33874135017395,
      "epoch": 0.6427497494153024,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004964792109234662,
      "loss": 5.3099,
      "mean_token_accuracy": 0.18377748429775237,
      "num_tokens": 14163749.0,
      "step": 7695
    },
    {
      "entropy": 5.312364149093628,
      "epoch": 0.6431673905780154,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004964739631258411,
      "loss": 5.23,
      "mean_token_accuracy": 0.18665992915630342,
      "num_tokens": 14174170.0,
      "step": 7700
    },
    {
      "entropy": 5.318781089782715,
      "epoch": 0.6435850317407283,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004964687114510478,
      "loss": 5.2222,
      "mean_token_accuracy": 0.18446299880743028,
      "num_tokens": 14182625.0,
      "step": 7705
    },
    {
      "entropy": 5.336656665802002,
      "epoch": 0.6440026729034414,
      "grad_norm": 1.34375,
      "learning_rate": 0.000496463455899178,
      "loss": 5.2138,
      "mean_token_accuracy": 0.18309882432222366,
      "num_tokens": 14192351.0,
      "step": 7710
    },
    {
      "entropy": 5.241625595092773,
      "epoch": 0.6444203140661544,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004964581964703239,
      "loss": 5.1285,
      "mean_token_accuracy": 0.18879476487636565,
      "num_tokens": 14201107.0,
      "step": 7715
    },
    {
      "entropy": 5.292919874191284,
      "epoch": 0.6448379552288673,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004964529331645776,
      "loss": 5.1796,
      "mean_token_accuracy": 0.18920957595109938,
      "num_tokens": 14209540.0,
      "step": 7720
    },
    {
      "entropy": 5.262211751937866,
      "epoch": 0.6452555963915804,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000496447665982031,
      "loss": 5.095,
      "mean_token_accuracy": 0.19543268382549286,
      "num_tokens": 14217944.0,
      "step": 7725
    },
    {
      "entropy": 5.246408271789551,
      "epoch": 0.6456732375542934,
      "grad_norm": 1.25,
      "learning_rate": 0.0004964423949227765,
      "loss": 5.1479,
      "mean_token_accuracy": 0.1906680718064308,
      "num_tokens": 14226044.0,
      "step": 7730
    },
    {
      "entropy": 5.333262348175049,
      "epoch": 0.6460908787170063,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004964371199869062,
      "loss": 5.2263,
      "mean_token_accuracy": 0.18008633106946945,
      "num_tokens": 14235566.0,
      "step": 7735
    },
    {
      "entropy": 5.31890230178833,
      "epoch": 0.6465085198797194,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004964318411745127,
      "loss": 5.1542,
      "mean_token_accuracy": 0.19815819561481476,
      "num_tokens": 14244511.0,
      "step": 7740
    },
    {
      "entropy": 5.269383287429809,
      "epoch": 0.6469261610424324,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004964265584856883,
      "loss": 5.2263,
      "mean_token_accuracy": 0.18629941046237947,
      "num_tokens": 14254299.0,
      "step": 7745
    },
    {
      "entropy": 5.215416622161865,
      "epoch": 0.6473438022051453,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004964212719205254,
      "loss": 5.1673,
      "mean_token_accuracy": 0.18904361575841905,
      "num_tokens": 14262915.0,
      "step": 7750
    },
    {
      "entropy": 5.3631199359893795,
      "epoch": 0.6477614433678583,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004964159814791167,
      "loss": 5.216,
      "mean_token_accuracy": 0.18888179510831832,
      "num_tokens": 14272606.0,
      "step": 7755
    },
    {
      "entropy": 5.312502145767212,
      "epoch": 0.6481790845305714,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004964106871615547,
      "loss": 5.2912,
      "mean_token_accuracy": 0.1826904833316803,
      "num_tokens": 14282117.0,
      "step": 7760
    },
    {
      "entropy": 5.247224283218384,
      "epoch": 0.6485967256932843,
      "grad_norm": 1.359375,
      "learning_rate": 0.000496405388967932,
      "loss": 5.2127,
      "mean_token_accuracy": 0.1859871491789818,
      "num_tokens": 14290421.0,
      "step": 7765
    },
    {
      "entropy": 5.446470260620117,
      "epoch": 0.6490143668559973,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004964000868983416,
      "loss": 5.3077,
      "mean_token_accuracy": 0.18256796002388,
      "num_tokens": 14299870.0,
      "step": 7770
    },
    {
      "entropy": 5.401399040222168,
      "epoch": 0.6494320080187104,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004963947809528761,
      "loss": 5.2342,
      "mean_token_accuracy": 0.18058783560991287,
      "num_tokens": 14309916.0,
      "step": 7775
    },
    {
      "entropy": 5.269759178161621,
      "epoch": 0.6498496491814233,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004963894711316285,
      "loss": 5.2175,
      "mean_token_accuracy": 0.18339438885450363,
      "num_tokens": 14319061.0,
      "step": 7780
    },
    {
      "entropy": 5.302542686462402,
      "epoch": 0.6502672903441363,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004963841574346917,
      "loss": 5.1677,
      "mean_token_accuracy": 0.19331397712230683,
      "num_tokens": 14327241.0,
      "step": 7785
    },
    {
      "entropy": 5.358416891098022,
      "epoch": 0.6506849315068494,
      "grad_norm": 1.25,
      "learning_rate": 0.0004963788398621587,
      "loss": 5.2916,
      "mean_token_accuracy": 0.1789262890815735,
      "num_tokens": 14336586.0,
      "step": 7790
    },
    {
      "entropy": 5.379330062866211,
      "epoch": 0.6511025726695623,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004963735184141227,
      "loss": 5.2489,
      "mean_token_accuracy": 0.18089451193809508,
      "num_tokens": 14346484.0,
      "step": 7795
    },
    {
      "entropy": 5.385251522064209,
      "epoch": 0.6515202138322753,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004963681930906768,
      "loss": 5.3219,
      "mean_token_accuracy": 0.17985379099845886,
      "num_tokens": 14356255.0,
      "step": 7800
    },
    {
      "entropy": 5.317829895019531,
      "epoch": 0.6519378549949884,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004963628638919141,
      "loss": 5.2595,
      "mean_token_accuracy": 0.18174924701452255,
      "num_tokens": 14365680.0,
      "step": 7805
    },
    {
      "entropy": 5.372163724899292,
      "epoch": 0.6523554961577013,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004963575308179281,
      "loss": 5.2288,
      "mean_token_accuracy": 0.19130903333425522,
      "num_tokens": 14374546.0,
      "step": 7810
    },
    {
      "entropy": 5.370970344543457,
      "epoch": 0.6527731373204143,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004963521938688122,
      "loss": 5.2405,
      "mean_token_accuracy": 0.1898658886551857,
      "num_tokens": 14383624.0,
      "step": 7815
    },
    {
      "entropy": 5.36236720085144,
      "epoch": 0.6531907784831273,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004963468530446596,
      "loss": 5.3447,
      "mean_token_accuracy": 0.17924101203680037,
      "num_tokens": 14392482.0,
      "step": 7820
    },
    {
      "entropy": 5.374128103256226,
      "epoch": 0.6536084196458403,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004963415083455639,
      "loss": 5.2385,
      "mean_token_accuracy": 0.18695005774497986,
      "num_tokens": 14401940.0,
      "step": 7825
    },
    {
      "entropy": 5.341295433044434,
      "epoch": 0.6540260608085533,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004963361597716186,
      "loss": 5.2989,
      "mean_token_accuracy": 0.18396753668785096,
      "num_tokens": 14411127.0,
      "step": 7830
    },
    {
      "entropy": 5.336658525466919,
      "epoch": 0.6544437019712663,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004963308073229175,
      "loss": 5.2092,
      "mean_token_accuracy": 0.18722108751535416,
      "num_tokens": 14421358.0,
      "step": 7835
    },
    {
      "entropy": 5.313114643096924,
      "epoch": 0.6548613431339793,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004963254509995542,
      "loss": 5.2262,
      "mean_token_accuracy": 0.18385069072246552,
      "num_tokens": 14430507.0,
      "step": 7840
    },
    {
      "entropy": 5.273292779922485,
      "epoch": 0.6552789842966923,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004963200908016224,
      "loss": 5.2111,
      "mean_token_accuracy": 0.19009640514850618,
      "num_tokens": 14439723.0,
      "step": 7845
    },
    {
      "entropy": 5.352698135375976,
      "epoch": 0.6556966254594053,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000496314726729216,
      "loss": 5.1857,
      "mean_token_accuracy": 0.18810579031705857,
      "num_tokens": 14449011.0,
      "step": 7850
    },
    {
      "entropy": 5.3103841781616214,
      "epoch": 0.6561142666221182,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004963093587824289,
      "loss": 5.1752,
      "mean_token_accuracy": 0.19130851477384567,
      "num_tokens": 14457676.0,
      "step": 7855
    },
    {
      "entropy": 5.342246437072754,
      "epoch": 0.6565319077848313,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004963039869613551,
      "loss": 5.2601,
      "mean_token_accuracy": 0.18594553023576738,
      "num_tokens": 14466956.0,
      "step": 7860
    },
    {
      "entropy": 5.313938426971435,
      "epoch": 0.6569495489475443,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004962986112660887,
      "loss": 5.1454,
      "mean_token_accuracy": 0.19567489624023438,
      "num_tokens": 14476083.0,
      "step": 7865
    },
    {
      "entropy": 5.365322732925415,
      "epoch": 0.6573671901102572,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004962932316967237,
      "loss": 5.2503,
      "mean_token_accuracy": 0.18117107152938844,
      "num_tokens": 14484980.0,
      "step": 7870
    },
    {
      "entropy": 5.355501651763916,
      "epoch": 0.6577848312729703,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004962878482533544,
      "loss": 5.2673,
      "mean_token_accuracy": 0.1856757715344429,
      "num_tokens": 14493965.0,
      "step": 7875
    },
    {
      "entropy": 5.350685262680054,
      "epoch": 0.6582024724356833,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004962824609360748,
      "loss": 5.2256,
      "mean_token_accuracy": 0.1830467164516449,
      "num_tokens": 14501914.0,
      "step": 7880
    },
    {
      "entropy": 5.239622974395752,
      "epoch": 0.6586201135983962,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004962770697449794,
      "loss": 5.1815,
      "mean_token_accuracy": 0.18943492770195008,
      "num_tokens": 14511447.0,
      "step": 7885
    },
    {
      "entropy": 5.301117324829102,
      "epoch": 0.6590377547611093,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004962716746801625,
      "loss": 5.1753,
      "mean_token_accuracy": 0.18255795240402223,
      "num_tokens": 14520558.0,
      "step": 7890
    },
    {
      "entropy": 5.291094636917114,
      "epoch": 0.6594553959238223,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004962662757417187,
      "loss": 5.1638,
      "mean_token_accuracy": 0.18984782993793486,
      "num_tokens": 14529995.0,
      "step": 7895
    },
    {
      "entropy": 5.3401213645935055,
      "epoch": 0.6598730370865352,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004962608729297424,
      "loss": 5.2979,
      "mean_token_accuracy": 0.18172093480825424,
      "num_tokens": 14538972.0,
      "step": 7900
    },
    {
      "entropy": 5.360935163497925,
      "epoch": 0.6602906782492483,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004962554662443281,
      "loss": 5.237,
      "mean_token_accuracy": 0.1828315943479538,
      "num_tokens": 14548255.0,
      "step": 7905
    },
    {
      "entropy": 5.304817485809326,
      "epoch": 0.6607083194119613,
      "grad_norm": 1.25,
      "learning_rate": 0.0004962500556855706,
      "loss": 5.1747,
      "mean_token_accuracy": 0.1892821729183197,
      "num_tokens": 14557265.0,
      "step": 7910
    },
    {
      "entropy": 5.350543355941772,
      "epoch": 0.6611259605746742,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004962446412535645,
      "loss": 5.2664,
      "mean_token_accuracy": 0.1839459255337715,
      "num_tokens": 14566650.0,
      "step": 7915
    },
    {
      "entropy": 5.32527379989624,
      "epoch": 0.6615436017373872,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004962392229484046,
      "loss": 5.1575,
      "mean_token_accuracy": 0.19818112701177598,
      "num_tokens": 14575550.0,
      "step": 7920
    },
    {
      "entropy": 5.297174882888794,
      "epoch": 0.6619612429001003,
      "grad_norm": 1.1875,
      "learning_rate": 0.000496233800770186,
      "loss": 5.1563,
      "mean_token_accuracy": 0.18891645818948746,
      "num_tokens": 14584459.0,
      "step": 7925
    },
    {
      "entropy": 5.371868562698364,
      "epoch": 0.6623788840628132,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004962283747190033,
      "loss": 5.2908,
      "mean_token_accuracy": 0.1820622742176056,
      "num_tokens": 14593709.0,
      "step": 7930
    },
    {
      "entropy": 5.253445720672607,
      "epoch": 0.6627965252255262,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004962229447949515,
      "loss": 5.1403,
      "mean_token_accuracy": 0.19019092470407487,
      "num_tokens": 14602071.0,
      "step": 7935
    },
    {
      "entropy": 5.417607975006104,
      "epoch": 0.6632141663882393,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004962175109981259,
      "loss": 5.3542,
      "mean_token_accuracy": 0.17659229636192322,
      "num_tokens": 14612842.0,
      "step": 7940
    },
    {
      "entropy": 5.464873266220093,
      "epoch": 0.6636318075509522,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004962120733286215,
      "loss": 5.3506,
      "mean_token_accuracy": 0.18645621240139007,
      "num_tokens": 14621645.0,
      "step": 7945
    },
    {
      "entropy": 5.294732666015625,
      "epoch": 0.6640494487136652,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004962066317865334,
      "loss": 5.1565,
      "mean_token_accuracy": 0.18930456191301345,
      "num_tokens": 14630447.0,
      "step": 7950
    },
    {
      "entropy": 5.258000373840332,
      "epoch": 0.6644670898763783,
      "grad_norm": 1.25,
      "learning_rate": 0.000496201186371957,
      "loss": 5.2554,
      "mean_token_accuracy": 0.19066265672445298,
      "num_tokens": 14639371.0,
      "step": 7955
    },
    {
      "entropy": 5.372070026397705,
      "epoch": 0.6648847310390912,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004961957370849875,
      "loss": 5.1465,
      "mean_token_accuracy": 0.18601731061935425,
      "num_tokens": 14648540.0,
      "step": 7960
    },
    {
      "entropy": 5.339339542388916,
      "epoch": 0.6653023722018042,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004961902839257205,
      "loss": 5.3118,
      "mean_token_accuracy": 0.18767533153295518,
      "num_tokens": 14658130.0,
      "step": 7965
    },
    {
      "entropy": 5.403484106063843,
      "epoch": 0.6657200133645172,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004961848268942513,
      "loss": 5.2674,
      "mean_token_accuracy": 0.1809846743941307,
      "num_tokens": 14667538.0,
      "step": 7970
    },
    {
      "entropy": 5.361376237869263,
      "epoch": 0.6661376545272302,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004961793659906754,
      "loss": 5.2359,
      "mean_token_accuracy": 0.18380988091230394,
      "num_tokens": 14677115.0,
      "step": 7975
    },
    {
      "entropy": 5.355688428878784,
      "epoch": 0.6665552956899432,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004961739012150885,
      "loss": 5.2136,
      "mean_token_accuracy": 0.19125464260578157,
      "num_tokens": 14686444.0,
      "step": 7980
    },
    {
      "entropy": 5.298751592636108,
      "epoch": 0.6669729368526562,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004961684325675862,
      "loss": 5.1785,
      "mean_token_accuracy": 0.19077081084251404,
      "num_tokens": 14695951.0,
      "step": 7985
    },
    {
      "entropy": 5.2982659339904785,
      "epoch": 0.6673905780153692,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004961629600482643,
      "loss": 5.3024,
      "mean_token_accuracy": 0.18561783134937287,
      "num_tokens": 14705415.0,
      "step": 7990
    },
    {
      "entropy": 5.424433374404908,
      "epoch": 0.6678082191780822,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004961574836572186,
      "loss": 5.2703,
      "mean_token_accuracy": 0.18185556530952454,
      "num_tokens": 14714793.0,
      "step": 7995
    },
    {
      "entropy": 5.209049034118652,
      "epoch": 0.6682258603407952,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004961520033945448,
      "loss": 5.0519,
      "mean_token_accuracy": 0.19594983905553817,
      "num_tokens": 14723820.0,
      "step": 8000
    },
    {
      "entropy": 5.244827556610107,
      "epoch": 0.6686435015035082,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004961465192603391,
      "loss": 5.1655,
      "mean_token_accuracy": 0.19469864219427108,
      "num_tokens": 14733055.0,
      "step": 8005
    },
    {
      "entropy": 5.334292984008789,
      "epoch": 0.6690611426662212,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004961410312546974,
      "loss": 5.1865,
      "mean_token_accuracy": 0.18282557129859925,
      "num_tokens": 14742416.0,
      "step": 8010
    },
    {
      "entropy": 5.26261534690857,
      "epoch": 0.6694787838289342,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004961355393777158,
      "loss": 5.2,
      "mean_token_accuracy": 0.1863041937351227,
      "num_tokens": 14751384.0,
      "step": 8015
    },
    {
      "entropy": 5.2433274269104,
      "epoch": 0.6698964249916471,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004961300436294902,
      "loss": 5.0799,
      "mean_token_accuracy": 0.18841362446546556,
      "num_tokens": 14760525.0,
      "step": 8020
    },
    {
      "entropy": 5.303926610946656,
      "epoch": 0.6703140661543602,
      "grad_norm": 1.5,
      "learning_rate": 0.0004961245440101171,
      "loss": 5.2477,
      "mean_token_accuracy": 0.18064588010311128,
      "num_tokens": 14769405.0,
      "step": 8025
    },
    {
      "entropy": 5.225086688995361,
      "epoch": 0.6707317073170732,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004961190405196928,
      "loss": 5.1147,
      "mean_token_accuracy": 0.19077132344245912,
      "num_tokens": 14778945.0,
      "step": 8030
    },
    {
      "entropy": 5.383445405960083,
      "epoch": 0.6711493484797861,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004961135331583135,
      "loss": 5.285,
      "mean_token_accuracy": 0.17766177356243135,
      "num_tokens": 14788289.0,
      "step": 8035
    },
    {
      "entropy": 5.292051124572754,
      "epoch": 0.6715669896424992,
      "grad_norm": 1.25,
      "learning_rate": 0.0004961080219260755,
      "loss": 5.2113,
      "mean_token_accuracy": 0.1816340357065201,
      "num_tokens": 14797761.0,
      "step": 8040
    },
    {
      "entropy": 5.374523687362671,
      "epoch": 0.6719846308052122,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004961025068230757,
      "loss": 5.2663,
      "mean_token_accuracy": 0.18736185282468795,
      "num_tokens": 14806424.0,
      "step": 8045
    },
    {
      "entropy": 5.294801330566406,
      "epoch": 0.6724022719679251,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004960969878494103,
      "loss": 5.1975,
      "mean_token_accuracy": 0.18426212519407273,
      "num_tokens": 14815816.0,
      "step": 8050
    },
    {
      "entropy": 5.279285097122193,
      "epoch": 0.6728199131306382,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004960914650051758,
      "loss": 5.1966,
      "mean_token_accuracy": 0.18834995329380036,
      "num_tokens": 14824877.0,
      "step": 8055
    },
    {
      "entropy": 5.23910698890686,
      "epoch": 0.6732375542933512,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004960859382904691,
      "loss": 5.0973,
      "mean_token_accuracy": 0.20258977860212327,
      "num_tokens": 14833095.0,
      "step": 8060
    },
    {
      "entropy": 5.363849925994873,
      "epoch": 0.6736551954560641,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000496080407705387,
      "loss": 5.2326,
      "mean_token_accuracy": 0.18313853591680526,
      "num_tokens": 14842963.0,
      "step": 8065
    },
    {
      "entropy": 5.38482666015625,
      "epoch": 0.6740728366187771,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004960748732500263,
      "loss": 5.3047,
      "mean_token_accuracy": 0.1805562898516655,
      "num_tokens": 14851552.0,
      "step": 8070
    },
    {
      "entropy": 5.3721260070800785,
      "epoch": 0.6744904777814902,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004960693349244838,
      "loss": 5.2082,
      "mean_token_accuracy": 0.19036448746919632,
      "num_tokens": 14859997.0,
      "step": 8075
    },
    {
      "entropy": 5.288872146606446,
      "epoch": 0.6749081189442031,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004960637927288566,
      "loss": 5.1288,
      "mean_token_accuracy": 0.18826107084751129,
      "num_tokens": 14868050.0,
      "step": 8080
    },
    {
      "entropy": 5.298485517501831,
      "epoch": 0.6753257601069161,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004960582466632414,
      "loss": 5.249,
      "mean_token_accuracy": 0.1802084356546402,
      "num_tokens": 14877774.0,
      "step": 8085
    },
    {
      "entropy": 5.231218576431274,
      "epoch": 0.6757434012696292,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004960526967277356,
      "loss": 5.1774,
      "mean_token_accuracy": 0.19676312804222107,
      "num_tokens": 14886971.0,
      "step": 8090
    },
    {
      "entropy": 5.2804406642913815,
      "epoch": 0.6761610424323421,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004960471429224363,
      "loss": 5.1199,
      "mean_token_accuracy": 0.19002149999141693,
      "num_tokens": 14895815.0,
      "step": 8095
    },
    {
      "entropy": 5.316021299362182,
      "epoch": 0.6765786835950551,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004960415852474406,
      "loss": 5.1514,
      "mean_token_accuracy": 0.19465539008378982,
      "num_tokens": 14904082.0,
      "step": 8100
    },
    {
      "entropy": 5.258999252319336,
      "epoch": 0.6769963247577682,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000496036023702846,
      "loss": 5.184,
      "mean_token_accuracy": 0.19023237526416778,
      "num_tokens": 14913533.0,
      "step": 8105
    },
    {
      "entropy": 5.273725986480713,
      "epoch": 0.6774139659204811,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004960304582887497,
      "loss": 5.1389,
      "mean_token_accuracy": 0.19112505614757538,
      "num_tokens": 14922063.0,
      "step": 8110
    },
    {
      "entropy": 5.309419727325439,
      "epoch": 0.6778316070831941,
      "grad_norm": 1.25,
      "learning_rate": 0.0004960248890052493,
      "loss": 5.1822,
      "mean_token_accuracy": 0.1919051840901375,
      "num_tokens": 14930219.0,
      "step": 8115
    },
    {
      "entropy": 5.348616456985473,
      "epoch": 0.6782492482459072,
      "grad_norm": 1.390625,
      "learning_rate": 0.000496019315852442,
      "loss": 5.2824,
      "mean_token_accuracy": 0.19079013168811798,
      "num_tokens": 14939463.0,
      "step": 8120
    },
    {
      "entropy": 5.394223928451538,
      "epoch": 0.6786668894086201,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004960137388304255,
      "loss": 5.1916,
      "mean_token_accuracy": 0.18611440658569336,
      "num_tokens": 14948289.0,
      "step": 8125
    },
    {
      "entropy": 5.355359029769898,
      "epoch": 0.6790845305713331,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004960081579392975,
      "loss": 5.2748,
      "mean_token_accuracy": 0.18862151056528093,
      "num_tokens": 14958172.0,
      "step": 8130
    },
    {
      "entropy": 5.220905733108521,
      "epoch": 0.6795021717340461,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004960025731791558,
      "loss": 5.078,
      "mean_token_accuracy": 0.19197088927030564,
      "num_tokens": 14967362.0,
      "step": 8135
    },
    {
      "entropy": 5.26541748046875,
      "epoch": 0.6799198128967591,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004959969845500979,
      "loss": 5.2696,
      "mean_token_accuracy": 0.18232544213533403,
      "num_tokens": 14975942.0,
      "step": 8140
    },
    {
      "entropy": 5.335718727111816,
      "epoch": 0.6803374540594721,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004959913920522218,
      "loss": 5.2111,
      "mean_token_accuracy": 0.1935685396194458,
      "num_tokens": 14985428.0,
      "step": 8145
    },
    {
      "entropy": 5.30684871673584,
      "epoch": 0.6807550952221851,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004959857956856254,
      "loss": 5.138,
      "mean_token_accuracy": 0.18895146995782852,
      "num_tokens": 14994558.0,
      "step": 8150
    },
    {
      "entropy": 5.306473255157471,
      "epoch": 0.6811727363848981,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004959801954504067,
      "loss": 5.1565,
      "mean_token_accuracy": 0.1863776907324791,
      "num_tokens": 15003674.0,
      "step": 8155
    },
    {
      "entropy": 5.312219905853271,
      "epoch": 0.6815903775476111,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004959745913466635,
      "loss": 5.2275,
      "mean_token_accuracy": 0.1866793751716614,
      "num_tokens": 15012569.0,
      "step": 8160
    },
    {
      "entropy": 5.343707275390625,
      "epoch": 0.6820080187103241,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004959689833744942,
      "loss": 5.2425,
      "mean_token_accuracy": 0.18308268189430238,
      "num_tokens": 15021863.0,
      "step": 8165
    },
    {
      "entropy": 5.386544704437256,
      "epoch": 0.682425659873037,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004959633715339969,
      "loss": 5.276,
      "mean_token_accuracy": 0.17785509824752807,
      "num_tokens": 15032272.0,
      "step": 8170
    },
    {
      "entropy": 5.320672607421875,
      "epoch": 0.6828433010357501,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004959577558252698,
      "loss": 5.197,
      "mean_token_accuracy": 0.17869272232055664,
      "num_tokens": 15041066.0,
      "step": 8175
    },
    {
      "entropy": 5.427869081497192,
      "epoch": 0.6832609421984631,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004959521362484112,
      "loss": 5.2962,
      "mean_token_accuracy": 0.17796089202165605,
      "num_tokens": 15050855.0,
      "step": 8180
    },
    {
      "entropy": 5.307919597625732,
      "epoch": 0.683678583361176,
      "grad_norm": 1.25,
      "learning_rate": 0.0004959465128035195,
      "loss": 5.1818,
      "mean_token_accuracy": 0.19280252903699874,
      "num_tokens": 15060810.0,
      "step": 8185
    },
    {
      "entropy": 5.296663665771485,
      "epoch": 0.6840962245238891,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004959408854906932,
      "loss": 5.1761,
      "mean_token_accuracy": 0.18804777413606644,
      "num_tokens": 15069761.0,
      "step": 8190
    },
    {
      "entropy": 5.285875606536865,
      "epoch": 0.6845138656866021,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004959352543100306,
      "loss": 5.1566,
      "mean_token_accuracy": 0.18793474435806273,
      "num_tokens": 15078691.0,
      "step": 8195
    },
    {
      "entropy": 5.234794330596924,
      "epoch": 0.684931506849315,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004959296192616305,
      "loss": 5.1452,
      "mean_token_accuracy": 0.18997769951820373,
      "num_tokens": 15087068.0,
      "step": 8200
    },
    {
      "entropy": 5.243432331085205,
      "epoch": 0.6853491480120281,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004959239803455916,
      "loss": 5.0311,
      "mean_token_accuracy": 0.1905503824353218,
      "num_tokens": 15095962.0,
      "step": 8205
    },
    {
      "entropy": 5.2430377960205075,
      "epoch": 0.6857667891747411,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004959183375620123,
      "loss": 5.2124,
      "mean_token_accuracy": 0.18324321359395981,
      "num_tokens": 15105835.0,
      "step": 8210
    },
    {
      "entropy": 5.3210914611816404,
      "epoch": 0.686184430337454,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004959126909109918,
      "loss": 5.1313,
      "mean_token_accuracy": 0.2020833745598793,
      "num_tokens": 15114140.0,
      "step": 8215
    },
    {
      "entropy": 5.332093524932861,
      "epoch": 0.686602071500167,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004959070403926286,
      "loss": 5.242,
      "mean_token_accuracy": 0.18745189607143403,
      "num_tokens": 15123253.0,
      "step": 8220
    },
    {
      "entropy": 5.424120044708252,
      "epoch": 0.6870197126628801,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004959013860070217,
      "loss": 5.3353,
      "mean_token_accuracy": 0.18029178977012633,
      "num_tokens": 15133699.0,
      "step": 8225
    },
    {
      "entropy": 5.366640901565551,
      "epoch": 0.687437353825593,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004958957277542701,
      "loss": 5.1054,
      "mean_token_accuracy": 0.19338543862104415,
      "num_tokens": 15142133.0,
      "step": 8230
    },
    {
      "entropy": 5.278004550933838,
      "epoch": 0.687854994988306,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004958900656344731,
      "loss": 5.1895,
      "mean_token_accuracy": 0.18659888803958893,
      "num_tokens": 15151367.0,
      "step": 8235
    },
    {
      "entropy": 5.278072834014893,
      "epoch": 0.6882726361510191,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004958843996477296,
      "loss": 5.2122,
      "mean_token_accuracy": 0.1813654512166977,
      "num_tokens": 15160938.0,
      "step": 8240
    },
    {
      "entropy": 5.383914279937744,
      "epoch": 0.688690277313732,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004958787297941386,
      "loss": 5.2512,
      "mean_token_accuracy": 0.18058507591485978,
      "num_tokens": 15170507.0,
      "step": 8245
    },
    {
      "entropy": 5.366903924942017,
      "epoch": 0.689107918476445,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004958730560737998,
      "loss": 5.2585,
      "mean_token_accuracy": 0.18802482038736343,
      "num_tokens": 15179822.0,
      "step": 8250
    },
    {
      "entropy": 5.277652025222778,
      "epoch": 0.6895255596391581,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004958673784868121,
      "loss": 5.1641,
      "mean_token_accuracy": 0.18544214367866516,
      "num_tokens": 15189353.0,
      "step": 8255
    },
    {
      "entropy": 5.358259677886963,
      "epoch": 0.689943200801871,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004958616970332752,
      "loss": 5.241,
      "mean_token_accuracy": 0.17864731103181838,
      "num_tokens": 15197772.0,
      "step": 8260
    },
    {
      "entropy": 5.255817794799805,
      "epoch": 0.690360841964584,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004958560117132884,
      "loss": 5.1884,
      "mean_token_accuracy": 0.19048339873552322,
      "num_tokens": 15207453.0,
      "step": 8265
    },
    {
      "entropy": 5.281562089920044,
      "epoch": 0.6907784831272971,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004958503225269513,
      "loss": 5.1082,
      "mean_token_accuracy": 0.19191768616437913,
      "num_tokens": 15215787.0,
      "step": 8270
    },
    {
      "entropy": 5.367995691299439,
      "epoch": 0.69119612429001,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004958446294743633,
      "loss": 5.2258,
      "mean_token_accuracy": 0.19048880338668822,
      "num_tokens": 15225148.0,
      "step": 8275
    },
    {
      "entropy": 5.309956502914429,
      "epoch": 0.691613765452723,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004958389325556245,
      "loss": 5.2556,
      "mean_token_accuracy": 0.18627925366163253,
      "num_tokens": 15234533.0,
      "step": 8280
    },
    {
      "entropy": 5.332719612121582,
      "epoch": 0.692031406615436,
      "grad_norm": 1.25,
      "learning_rate": 0.0004958332317708341,
      "loss": 5.2605,
      "mean_token_accuracy": 0.1786329224705696,
      "num_tokens": 15243805.0,
      "step": 8285
    },
    {
      "entropy": 5.406910753250122,
      "epoch": 0.692449047778149,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004958275271200924,
      "loss": 5.2691,
      "mean_token_accuracy": 0.1810326337814331,
      "num_tokens": 15253298.0,
      "step": 8290
    },
    {
      "entropy": 5.297081518173218,
      "epoch": 0.692866688940862,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004958218186034988,
      "loss": 5.2197,
      "mean_token_accuracy": 0.18782705962657928,
      "num_tokens": 15261860.0,
      "step": 8295
    },
    {
      "entropy": 5.288454246520996,
      "epoch": 0.693284330103575,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004958161062211537,
      "loss": 5.1586,
      "mean_token_accuracy": 0.18947628289461135,
      "num_tokens": 15271021.0,
      "step": 8300
    },
    {
      "entropy": 5.286473751068115,
      "epoch": 0.693701971266288,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004958103899731566,
      "loss": 5.1834,
      "mean_token_accuracy": 0.18143957555294038,
      "num_tokens": 15280489.0,
      "step": 8305
    },
    {
      "entropy": 5.342685127258301,
      "epoch": 0.694119612429001,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004958046698596081,
      "loss": 5.2105,
      "mean_token_accuracy": 0.19078896790742875,
      "num_tokens": 15289387.0,
      "step": 8310
    },
    {
      "entropy": 5.334821605682373,
      "epoch": 0.694537253591714,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004957989458806079,
      "loss": 5.27,
      "mean_token_accuracy": 0.18353917002677916,
      "num_tokens": 15299334.0,
      "step": 8315
    },
    {
      "entropy": 5.3309906959533695,
      "epoch": 0.694954894754427,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004957932180362564,
      "loss": 5.2408,
      "mean_token_accuracy": 0.18631087243556976,
      "num_tokens": 15309174.0,
      "step": 8320
    },
    {
      "entropy": 5.404925775527954,
      "epoch": 0.69537253591714,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004957874863266538,
      "loss": 5.2994,
      "mean_token_accuracy": 0.181829771399498,
      "num_tokens": 15318722.0,
      "step": 8325
    },
    {
      "entropy": 5.432412815093994,
      "epoch": 0.695790177079853,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004957817507519006,
      "loss": 5.2198,
      "mean_token_accuracy": 0.18473161309957503,
      "num_tokens": 15327606.0,
      "step": 8330
    },
    {
      "entropy": 5.250016069412231,
      "epoch": 0.696207818242566,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004957760113120971,
      "loss": 5.2082,
      "mean_token_accuracy": 0.19313850551843642,
      "num_tokens": 15336520.0,
      "step": 8335
    },
    {
      "entropy": 5.380140256881714,
      "epoch": 0.696625459405279,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004957702680073437,
      "loss": 5.2538,
      "mean_token_accuracy": 0.18560430109500886,
      "num_tokens": 15345437.0,
      "step": 8340
    },
    {
      "entropy": 5.393169927597046,
      "epoch": 0.697043100567992,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004957645208377411,
      "loss": 5.1909,
      "mean_token_accuracy": 0.1888318419456482,
      "num_tokens": 15354660.0,
      "step": 8345
    },
    {
      "entropy": 5.296236753463745,
      "epoch": 0.6974607417307049,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004957587698033898,
      "loss": 5.2369,
      "mean_token_accuracy": 0.19031170308589934,
      "num_tokens": 15363958.0,
      "step": 8350
    },
    {
      "entropy": 5.281468343734741,
      "epoch": 0.697878382893418,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004957530149043905,
      "loss": 5.1846,
      "mean_token_accuracy": 0.19381961673498155,
      "num_tokens": 15373351.0,
      "step": 8355
    },
    {
      "entropy": 5.363962316513062,
      "epoch": 0.698296024056131,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004957472561408438,
      "loss": 5.1906,
      "mean_token_accuracy": 0.1902700036764145,
      "num_tokens": 15383069.0,
      "step": 8360
    },
    {
      "entropy": 5.311092233657837,
      "epoch": 0.6987136652188439,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004957414935128509,
      "loss": 5.1815,
      "mean_token_accuracy": 0.18567148745059966,
      "num_tokens": 15391394.0,
      "step": 8365
    },
    {
      "entropy": 5.295946455001831,
      "epoch": 0.699131306381557,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004957357270205123,
      "loss": 5.1678,
      "mean_token_accuracy": 0.18772562444210053,
      "num_tokens": 15399993.0,
      "step": 8370
    },
    {
      "entropy": 5.244115734100342,
      "epoch": 0.69954894754427,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004957299566639292,
      "loss": 5.1766,
      "mean_token_accuracy": 0.18961432129144667,
      "num_tokens": 15409886.0,
      "step": 8375
    },
    {
      "entropy": 5.371616411209106,
      "epoch": 0.6999665887069829,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004957241824432025,
      "loss": 5.1691,
      "mean_token_accuracy": 0.19355978816747665,
      "num_tokens": 15419772.0,
      "step": 8380
    },
    {
      "entropy": 5.338627195358276,
      "epoch": 0.700384229869696,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004957184043584333,
      "loss": 5.1822,
      "mean_token_accuracy": 0.18718047738075255,
      "num_tokens": 15429631.0,
      "step": 8385
    },
    {
      "entropy": 5.225720691680908,
      "epoch": 0.700801871032409,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004957126224097227,
      "loss": 5.1339,
      "mean_token_accuracy": 0.19838509559631348,
      "num_tokens": 15438411.0,
      "step": 8390
    },
    {
      "entropy": 5.154391050338745,
      "epoch": 0.7012195121951219,
      "grad_norm": 1.15625,
      "learning_rate": 0.000495706836597172,
      "loss": 5.1628,
      "mean_token_accuracy": 0.1888429969549179,
      "num_tokens": 15448943.0,
      "step": 8395
    },
    {
      "entropy": 5.331808376312256,
      "epoch": 0.7016371533578349,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004957010469208825,
      "loss": 5.1135,
      "mean_token_accuracy": 0.19145280718803406,
      "num_tokens": 15457908.0,
      "step": 8400
    },
    {
      "entropy": 5.398460102081299,
      "epoch": 0.702054794520548,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004956952533809554,
      "loss": 5.2188,
      "mean_token_accuracy": 0.18720683008432387,
      "num_tokens": 15467390.0,
      "step": 8405
    },
    {
      "entropy": 5.275925016403198,
      "epoch": 0.7024724356832609,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004956894559774923,
      "loss": 5.2138,
      "mean_token_accuracy": 0.18277271687984467,
      "num_tokens": 15476419.0,
      "step": 8410
    },
    {
      "entropy": 5.192082023620605,
      "epoch": 0.7028900768459739,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004956836547105947,
      "loss": 5.0608,
      "mean_token_accuracy": 0.1960819587111473,
      "num_tokens": 15485147.0,
      "step": 8415
    },
    {
      "entropy": 5.358620834350586,
      "epoch": 0.703307718008687,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004956778495803641,
      "loss": 5.1752,
      "mean_token_accuracy": 0.19049345403909684,
      "num_tokens": 15493816.0,
      "step": 8420
    },
    {
      "entropy": 5.339207649230957,
      "epoch": 0.7037253591713999,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000495672040586902,
      "loss": 5.2998,
      "mean_token_accuracy": 0.18781712353229524,
      "num_tokens": 15503051.0,
      "step": 8425
    },
    {
      "entropy": 5.276934671401977,
      "epoch": 0.7041430003341129,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004956662277303103,
      "loss": 5.1023,
      "mean_token_accuracy": 0.19193120747804643,
      "num_tokens": 15511854.0,
      "step": 8430
    },
    {
      "entropy": 5.303779411315918,
      "epoch": 0.704560641496826,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004956604110106905,
      "loss": 5.2238,
      "mean_token_accuracy": 0.1936649650335312,
      "num_tokens": 15521492.0,
      "step": 8435
    },
    {
      "entropy": 5.428117752075195,
      "epoch": 0.7049782826595389,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004956545904281447,
      "loss": 5.3095,
      "mean_token_accuracy": 0.17229399383068084,
      "num_tokens": 15530983.0,
      "step": 8440
    },
    {
      "entropy": 5.332515287399292,
      "epoch": 0.7053959238222519,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004956487659827748,
      "loss": 5.1461,
      "mean_token_accuracy": 0.19503338932991027,
      "num_tokens": 15539722.0,
      "step": 8445
    },
    {
      "entropy": 5.223224401473999,
      "epoch": 0.705813564984965,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004956429376746825,
      "loss": 5.1259,
      "mean_token_accuracy": 0.1945224493741989,
      "num_tokens": 15548738.0,
      "step": 8450
    },
    {
      "entropy": 5.180972003936768,
      "epoch": 0.7062312061476779,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004956371055039701,
      "loss": 5.1517,
      "mean_token_accuracy": 0.1971757486462593,
      "num_tokens": 15557520.0,
      "step": 8455
    },
    {
      "entropy": 5.419539213180542,
      "epoch": 0.7066488473103909,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004956312694707395,
      "loss": 5.2405,
      "mean_token_accuracy": 0.1760064870119095,
      "num_tokens": 15567860.0,
      "step": 8460
    },
    {
      "entropy": 5.392957305908203,
      "epoch": 0.7070664884731039,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000495625429575093,
      "loss": 5.2296,
      "mean_token_accuracy": 0.18215364515781401,
      "num_tokens": 15577174.0,
      "step": 8465
    },
    {
      "entropy": 5.215926551818848,
      "epoch": 0.7074841296358169,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004956195858171327,
      "loss": 5.0285,
      "mean_token_accuracy": 0.19613219201564788,
      "num_tokens": 15585476.0,
      "step": 8470
    },
    {
      "entropy": 5.205816268920898,
      "epoch": 0.7079017707985299,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004956137381969611,
      "loss": 5.1776,
      "mean_token_accuracy": 0.19150867611169814,
      "num_tokens": 15595422.0,
      "step": 8475
    },
    {
      "entropy": 5.343913602828979,
      "epoch": 0.7083194119612429,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004956078867146805,
      "loss": 5.2868,
      "mean_token_accuracy": 0.18171995729207993,
      "num_tokens": 15604223.0,
      "step": 8480
    },
    {
      "entropy": 5.414848661422729,
      "epoch": 0.7087370531239559,
      "grad_norm": 1.25,
      "learning_rate": 0.0004956020313703933,
      "loss": 5.264,
      "mean_token_accuracy": 0.18618287444114684,
      "num_tokens": 15613715.0,
      "step": 8485
    },
    {
      "entropy": 5.268935060501098,
      "epoch": 0.7091546942866689,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004955961721642018,
      "loss": 5.0818,
      "mean_token_accuracy": 0.20129425972700118,
      "num_tokens": 15623826.0,
      "step": 8490
    },
    {
      "entropy": 5.25658106803894,
      "epoch": 0.7095723354493819,
      "grad_norm": 1.234375,
      "learning_rate": 0.000495590309096209,
      "loss": 5.1979,
      "mean_token_accuracy": 0.1842020645737648,
      "num_tokens": 15633311.0,
      "step": 8495
    },
    {
      "entropy": 5.3058939456939695,
      "epoch": 0.7099899766120948,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004955844421665171,
      "loss": 5.1821,
      "mean_token_accuracy": 0.19040406942367555,
      "num_tokens": 15642785.0,
      "step": 8500
    },
    {
      "entropy": 5.462882614135742,
      "epoch": 0.7104076177748079,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004955785713752292,
      "loss": 5.3747,
      "mean_token_accuracy": 0.18012833893299102,
      "num_tokens": 15653009.0,
      "step": 8505
    },
    {
      "entropy": 5.327106094360351,
      "epoch": 0.7108252589375209,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004955726967224479,
      "loss": 5.1778,
      "mean_token_accuracy": 0.19191291481256484,
      "num_tokens": 15661672.0,
      "step": 8510
    },
    {
      "entropy": 5.248064231872559,
      "epoch": 0.7112429001002338,
      "grad_norm": 1.28125,
      "learning_rate": 0.000495566818208276,
      "loss": 5.166,
      "mean_token_accuracy": 0.1860044613480568,
      "num_tokens": 15672186.0,
      "step": 8515
    },
    {
      "entropy": 5.253706312179565,
      "epoch": 0.7116605412629469,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004955609358328163,
      "loss": 5.182,
      "mean_token_accuracy": 0.19245645850896836,
      "num_tokens": 15680351.0,
      "step": 8520
    },
    {
      "entropy": 5.318135833740234,
      "epoch": 0.7120781824256599,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004955550495961722,
      "loss": 5.1784,
      "mean_token_accuracy": 0.1842174783349037,
      "num_tokens": 15690140.0,
      "step": 8525
    },
    {
      "entropy": 5.296385097503662,
      "epoch": 0.7124958235883728,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004955491594984462,
      "loss": 5.2574,
      "mean_token_accuracy": 0.18273524791002274,
      "num_tokens": 15699968.0,
      "step": 8530
    },
    {
      "entropy": 5.293054294586182,
      "epoch": 0.7129134647510859,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004955432655397419,
      "loss": 5.1124,
      "mean_token_accuracy": 0.1904730960726738,
      "num_tokens": 15709061.0,
      "step": 8535
    },
    {
      "entropy": 5.354221391677856,
      "epoch": 0.7133311059137989,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004955373677201621,
      "loss": 5.2486,
      "mean_token_accuracy": 0.18385081142187118,
      "num_tokens": 15718350.0,
      "step": 8540
    },
    {
      "entropy": 5.291756105422974,
      "epoch": 0.7137487470765118,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004955314660398105,
      "loss": 5.1663,
      "mean_token_accuracy": 0.1910427913069725,
      "num_tokens": 15727120.0,
      "step": 8545
    },
    {
      "entropy": 5.2999427795410154,
      "epoch": 0.7141663882392248,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004955255604987899,
      "loss": 5.2569,
      "mean_token_accuracy": 0.18272571116685868,
      "num_tokens": 15735865.0,
      "step": 8550
    },
    {
      "entropy": 5.2988136291503904,
      "epoch": 0.7145840294019379,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000495519651097204,
      "loss": 5.2018,
      "mean_token_accuracy": 0.18357786238193513,
      "num_tokens": 15745239.0,
      "step": 8555
    },
    {
      "entropy": 5.281273698806762,
      "epoch": 0.7150016705646508,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004955137378351561,
      "loss": 5.1311,
      "mean_token_accuracy": 0.19101595878601074,
      "num_tokens": 15754418.0,
      "step": 8560
    },
    {
      "entropy": 5.2618578433990475,
      "epoch": 0.7154193117273638,
      "grad_norm": 1.25,
      "learning_rate": 0.0004955078207127498,
      "loss": 5.1739,
      "mean_token_accuracy": 0.194094018638134,
      "num_tokens": 15764175.0,
      "step": 8565
    },
    {
      "entropy": 5.332169723510742,
      "epoch": 0.7158369528900769,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004955018997300886,
      "loss": 5.1738,
      "mean_token_accuracy": 0.18773175030946732,
      "num_tokens": 15772766.0,
      "step": 8570
    },
    {
      "entropy": 5.241567802429199,
      "epoch": 0.7162545940527898,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004954959748872764,
      "loss": 5.1854,
      "mean_token_accuracy": 0.19543151706457138,
      "num_tokens": 15781768.0,
      "step": 8575
    },
    {
      "entropy": 5.342124986648559,
      "epoch": 0.7166722352155028,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004954900461844167,
      "loss": 5.2192,
      "mean_token_accuracy": 0.18850684762001038,
      "num_tokens": 15791706.0,
      "step": 8580
    },
    {
      "entropy": 5.349425840377807,
      "epoch": 0.7170898763782159,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004954841136216132,
      "loss": 5.2588,
      "mean_token_accuracy": 0.18596038669347764,
      "num_tokens": 15802015.0,
      "step": 8585
    },
    {
      "entropy": 5.327006816864014,
      "epoch": 0.7175075175409288,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004954781771989701,
      "loss": 5.2005,
      "mean_token_accuracy": 0.18541175425052642,
      "num_tokens": 15811116.0,
      "step": 8590
    },
    {
      "entropy": 5.367788982391358,
      "epoch": 0.7179251587036418,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000495472236916591,
      "loss": 5.1679,
      "mean_token_accuracy": 0.18778144270181657,
      "num_tokens": 15819240.0,
      "step": 8595
    },
    {
      "entropy": 5.216233777999878,
      "epoch": 0.7183427998663549,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004954662927745799,
      "loss": 5.1806,
      "mean_token_accuracy": 0.18650583773851395,
      "num_tokens": 15827740.0,
      "step": 8600
    },
    {
      "entropy": 5.233916473388672,
      "epoch": 0.7187604410290678,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004954603447730411,
      "loss": 5.2334,
      "mean_token_accuracy": 0.18810488432645797,
      "num_tokens": 15836763.0,
      "step": 8605
    },
    {
      "entropy": 5.415298748016357,
      "epoch": 0.7191780821917808,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004954543929120786,
      "loss": 5.2872,
      "mean_token_accuracy": 0.18467582166194915,
      "num_tokens": 15846624.0,
      "step": 8610
    },
    {
      "entropy": 5.328110313415527,
      "epoch": 0.7195957233544938,
      "grad_norm": 1.25,
      "learning_rate": 0.0004954484371917967,
      "loss": 5.208,
      "mean_token_accuracy": 0.18547792434692384,
      "num_tokens": 15855713.0,
      "step": 8615
    },
    {
      "entropy": 5.257599639892578,
      "epoch": 0.7200133645172068,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004954424776122993,
      "loss": 5.0396,
      "mean_token_accuracy": 0.197099906206131,
      "num_tokens": 15863577.0,
      "step": 8620
    },
    {
      "entropy": 5.3231040954589846,
      "epoch": 0.7204310056799198,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004954365141736913,
      "loss": 5.2085,
      "mean_token_accuracy": 0.19211025387048722,
      "num_tokens": 15872479.0,
      "step": 8625
    },
    {
      "entropy": 5.356935358047485,
      "epoch": 0.7208486468426328,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004954305468760765,
      "loss": 5.2999,
      "mean_token_accuracy": 0.18382661193609237,
      "num_tokens": 15882170.0,
      "step": 8630
    },
    {
      "entropy": 5.27643313407898,
      "epoch": 0.7212662880053458,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004954245757195598,
      "loss": 5.1797,
      "mean_token_accuracy": 0.1908285766839981,
      "num_tokens": 15891694.0,
      "step": 8635
    },
    {
      "entropy": 5.323697233200074,
      "epoch": 0.7216839291680588,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004954186007042454,
      "loss": 5.2997,
      "mean_token_accuracy": 0.17307595759630204,
      "num_tokens": 15902232.0,
      "step": 8640
    },
    {
      "entropy": 5.364664793014526,
      "epoch": 0.7221015703307718,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004954126218302382,
      "loss": 5.1564,
      "mean_token_accuracy": 0.1835283249616623,
      "num_tokens": 15912218.0,
      "step": 8645
    },
    {
      "entropy": 5.274173355102539,
      "epoch": 0.7225192114934847,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004954066390976427,
      "loss": 5.1159,
      "mean_token_accuracy": 0.19668061137199402,
      "num_tokens": 15922011.0,
      "step": 8650
    },
    {
      "entropy": 5.144522762298584,
      "epoch": 0.7229368526561978,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004954006525065639,
      "loss": 5.1051,
      "mean_token_accuracy": 0.1925290733575821,
      "num_tokens": 15930394.0,
      "step": 8655
    },
    {
      "entropy": 5.275206995010376,
      "epoch": 0.7233544938189108,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004953946620571061,
      "loss": 5.2164,
      "mean_token_accuracy": 0.1811532571911812,
      "num_tokens": 15939880.0,
      "step": 8660
    },
    {
      "entropy": 5.266812944412232,
      "epoch": 0.7237721349816237,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004953886677493746,
      "loss": 5.0984,
      "mean_token_accuracy": 0.19883121699094772,
      "num_tokens": 15948521.0,
      "step": 8665
    },
    {
      "entropy": 5.312144136428833,
      "epoch": 0.7241897761443368,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004953826695834741,
      "loss": 5.2031,
      "mean_token_accuracy": 0.18478854596614838,
      "num_tokens": 15956811.0,
      "step": 8670
    },
    {
      "entropy": 5.2990309715271,
      "epoch": 0.7246074173070498,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004953766675595098,
      "loss": 5.2462,
      "mean_token_accuracy": 0.19069269746541978,
      "num_tokens": 15967019.0,
      "step": 8675
    },
    {
      "entropy": 5.421168422698974,
      "epoch": 0.7250250584697627,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004953706616775866,
      "loss": 5.2299,
      "mean_token_accuracy": 0.18662516474723817,
      "num_tokens": 15976149.0,
      "step": 8680
    },
    {
      "entropy": 5.3284658908844,
      "epoch": 0.7254426996324758,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004953646519378097,
      "loss": 5.0604,
      "mean_token_accuracy": 0.1943461775779724,
      "num_tokens": 15986711.0,
      "step": 8685
    },
    {
      "entropy": 5.249668121337891,
      "epoch": 0.7258603407951888,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004953586383402844,
      "loss": 5.2049,
      "mean_token_accuracy": 0.18818472623825072,
      "num_tokens": 15996482.0,
      "step": 8690
    },
    {
      "entropy": 5.195931673049927,
      "epoch": 0.7262779819579017,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004953526208851157,
      "loss": 5.0965,
      "mean_token_accuracy": 0.18975579589605332,
      "num_tokens": 16005801.0,
      "step": 8695
    },
    {
      "entropy": 5.218811559677124,
      "epoch": 0.7266956231206148,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004953465995724092,
      "loss": 5.1268,
      "mean_token_accuracy": 0.18776528388261796,
      "num_tokens": 16014807.0,
      "step": 8700
    },
    {
      "entropy": 5.366499328613282,
      "epoch": 0.7271132642833278,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004953405744022705,
      "loss": 5.2722,
      "mean_token_accuracy": 0.18139951825141906,
      "num_tokens": 16023326.0,
      "step": 8705
    },
    {
      "entropy": 5.420980024337768,
      "epoch": 0.7275309054460407,
      "grad_norm": 1.25,
      "learning_rate": 0.0004953345453748045,
      "loss": 5.2269,
      "mean_token_accuracy": 0.18403305411338805,
      "num_tokens": 16031648.0,
      "step": 8710
    },
    {
      "entropy": 5.277630138397217,
      "epoch": 0.7279485466087537,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004953285124901171,
      "loss": 5.1451,
      "mean_token_accuracy": 0.19288977831602097,
      "num_tokens": 16040179.0,
      "step": 8715
    },
    {
      "entropy": 5.287181758880616,
      "epoch": 0.7283661877714668,
      "grad_norm": 1.1875,
      "learning_rate": 0.000495322475748314,
      "loss": 5.3103,
      "mean_token_accuracy": 0.1791713386774063,
      "num_tokens": 16050770.0,
      "step": 8720
    },
    {
      "entropy": 5.381537771224975,
      "epoch": 0.7287838289341797,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004953164351495007,
      "loss": 5.1633,
      "mean_token_accuracy": 0.1909000188112259,
      "num_tokens": 16060504.0,
      "step": 8725
    },
    {
      "entropy": 5.243673849105835,
      "epoch": 0.7292014700968927,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004953103906937831,
      "loss": 5.0057,
      "mean_token_accuracy": 0.20394542068243027,
      "num_tokens": 16069240.0,
      "step": 8730
    },
    {
      "entropy": 5.183630323410034,
      "epoch": 0.7296191112596058,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004953043423812669,
      "loss": 5.2643,
      "mean_token_accuracy": 0.18672844469547273,
      "num_tokens": 16079098.0,
      "step": 8735
    },
    {
      "entropy": 5.208294630050659,
      "epoch": 0.7300367524223187,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004952982902120579,
      "loss": 5.0756,
      "mean_token_accuracy": 0.19283230602741241,
      "num_tokens": 16088454.0,
      "step": 8740
    },
    {
      "entropy": 5.339796686172486,
      "epoch": 0.7304543935850317,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004952922341862622,
      "loss": 5.227,
      "mean_token_accuracy": 0.18357341587543488,
      "num_tokens": 16096985.0,
      "step": 8745
    },
    {
      "entropy": 5.436058807373047,
      "epoch": 0.7308720347477448,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004952861743039858,
      "loss": 5.3296,
      "mean_token_accuracy": 0.18211542069911957,
      "num_tokens": 16105434.0,
      "step": 8750
    },
    {
      "entropy": 5.225539636611939,
      "epoch": 0.7312896759104577,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004952801105653347,
      "loss": 5.0694,
      "mean_token_accuracy": 0.2016468346118927,
      "num_tokens": 16114933.0,
      "step": 8755
    },
    {
      "entropy": 5.321869230270385,
      "epoch": 0.7317073170731707,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004952740429704152,
      "loss": 5.2304,
      "mean_token_accuracy": 0.18201682567596436,
      "num_tokens": 16124954.0,
      "step": 8760
    },
    {
      "entropy": 5.265962553024292,
      "epoch": 0.7321249582358837,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004952679715193334,
      "loss": 5.0491,
      "mean_token_accuracy": 0.19678904116153717,
      "num_tokens": 16134442.0,
      "step": 8765
    },
    {
      "entropy": 5.26953535079956,
      "epoch": 0.7325425993985967,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004952618962121956,
      "loss": 5.23,
      "mean_token_accuracy": 0.1834096133708954,
      "num_tokens": 16143994.0,
      "step": 8770
    },
    {
      "entropy": 5.279685401916504,
      "epoch": 0.7329602405613097,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004952558170491082,
      "loss": 5.1388,
      "mean_token_accuracy": 0.18969169557094573,
      "num_tokens": 16152931.0,
      "step": 8775
    },
    {
      "entropy": 5.301347398757935,
      "epoch": 0.7333778817240227,
      "grad_norm": 1.375,
      "learning_rate": 0.0004952497340301776,
      "loss": 5.1601,
      "mean_token_accuracy": 0.1868045687675476,
      "num_tokens": 16162716.0,
      "step": 8780
    },
    {
      "entropy": 5.362487745285034,
      "epoch": 0.7337955228867357,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004952436471555103,
      "loss": 5.2116,
      "mean_token_accuracy": 0.18239102512598038,
      "num_tokens": 16172371.0,
      "step": 8785
    },
    {
      "entropy": 5.304409694671631,
      "epoch": 0.7342131640494487,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004952375564252128,
      "loss": 5.291,
      "mean_token_accuracy": 0.1842018783092499,
      "num_tokens": 16181823.0,
      "step": 8790
    },
    {
      "entropy": 5.339714431762696,
      "epoch": 0.7346308052121617,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004952314618393919,
      "loss": 5.1829,
      "mean_token_accuracy": 0.1916033521294594,
      "num_tokens": 16190720.0,
      "step": 8795
    },
    {
      "entropy": 5.296022081375122,
      "epoch": 0.7350484463748747,
      "grad_norm": 1.5546875,
      "learning_rate": 0.000495225363398154,
      "loss": 5.1291,
      "mean_token_accuracy": 0.1905677482485771,
      "num_tokens": 16199589.0,
      "step": 8800
    },
    {
      "entropy": 5.387620830535889,
      "epoch": 0.7354660875375877,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004952192611016061,
      "loss": 5.301,
      "mean_token_accuracy": 0.18462697565555572,
      "num_tokens": 16208800.0,
      "step": 8805
    },
    {
      "entropy": 5.334796905517578,
      "epoch": 0.7358837287003007,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000495213154949855,
      "loss": 5.2731,
      "mean_token_accuracy": 0.18237655311822892,
      "num_tokens": 16218835.0,
      "step": 8810
    },
    {
      "entropy": 5.3320722579956055,
      "epoch": 0.7363013698630136,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004952070449430075,
      "loss": 5.2247,
      "mean_token_accuracy": 0.18262556791305543,
      "num_tokens": 16228441.0,
      "step": 8815
    },
    {
      "entropy": 5.375856208801269,
      "epoch": 0.7367190110257267,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004952009310811706,
      "loss": 5.2163,
      "mean_token_accuracy": 0.19435302913188934,
      "num_tokens": 16237877.0,
      "step": 8820
    },
    {
      "entropy": 5.288750648498535,
      "epoch": 0.7371366521884397,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004951948133644514,
      "loss": 5.1154,
      "mean_token_accuracy": 0.18971776962280273,
      "num_tokens": 16246809.0,
      "step": 8825
    },
    {
      "entropy": 5.299361896514893,
      "epoch": 0.7375542933511526,
      "grad_norm": 1.328125,
      "learning_rate": 0.000495188691792957,
      "loss": 5.2135,
      "mean_token_accuracy": 0.19305830895900727,
      "num_tokens": 16255754.0,
      "step": 8830
    },
    {
      "entropy": 5.350617551803589,
      "epoch": 0.7379719345138657,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004951825663667943,
      "loss": 5.2233,
      "mean_token_accuracy": 0.1851801112294197,
      "num_tokens": 16264861.0,
      "step": 8835
    },
    {
      "entropy": 5.352020931243897,
      "epoch": 0.7383895756765787,
      "grad_norm": 1.203125,
      "learning_rate": 0.000495176437086071,
      "loss": 5.1803,
      "mean_token_accuracy": 0.18862801939249038,
      "num_tokens": 16273628.0,
      "step": 8840
    },
    {
      "entropy": 5.310912990570069,
      "epoch": 0.7388072168392917,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000495170303950894,
      "loss": 5.1744,
      "mean_token_accuracy": 0.18131401985883713,
      "num_tokens": 16282958.0,
      "step": 8845
    },
    {
      "entropy": 5.35797529220581,
      "epoch": 0.7392248580020047,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004951641669613708,
      "loss": 5.2065,
      "mean_token_accuracy": 0.18231176137924193,
      "num_tokens": 16292763.0,
      "step": 8850
    },
    {
      "entropy": 5.324119424819946,
      "epoch": 0.7396424991647177,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004951580261176088,
      "loss": 5.214,
      "mean_token_accuracy": 0.18999577313661575,
      "num_tokens": 16302216.0,
      "step": 8855
    },
    {
      "entropy": 5.216421508789063,
      "epoch": 0.7400601403274307,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004951518814197156,
      "loss": 5.1016,
      "mean_token_accuracy": 0.1944451928138733,
      "num_tokens": 16311672.0,
      "step": 8860
    },
    {
      "entropy": 5.310057544708252,
      "epoch": 0.7404777814901436,
      "grad_norm": 1.25,
      "learning_rate": 0.0004951457328677989,
      "loss": 5.118,
      "mean_token_accuracy": 0.1860945329070091,
      "num_tokens": 16321027.0,
      "step": 8865
    },
    {
      "entropy": 5.338929748535156,
      "epoch": 0.7408954226528567,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004951395804619659,
      "loss": 5.2232,
      "mean_token_accuracy": 0.1821383833885193,
      "num_tokens": 16329471.0,
      "step": 8870
    },
    {
      "entropy": 5.270784616470337,
      "epoch": 0.7413130638155697,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004951334242023247,
      "loss": 5.1862,
      "mean_token_accuracy": 0.18740254193544387,
      "num_tokens": 16339085.0,
      "step": 8875
    },
    {
      "entropy": 5.4150670051574705,
      "epoch": 0.7417307049782826,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004951272640889828,
      "loss": 5.253,
      "mean_token_accuracy": 0.1961119741201401,
      "num_tokens": 16347950.0,
      "step": 8880
    },
    {
      "entropy": 5.275183629989624,
      "epoch": 0.7421483461409957,
      "grad_norm": 1.375,
      "learning_rate": 0.0004951211001220484,
      "loss": 5.0294,
      "mean_token_accuracy": 0.1989552140235901,
      "num_tokens": 16357129.0,
      "step": 8885
    },
    {
      "entropy": 5.217408847808838,
      "epoch": 0.7425659873037087,
      "grad_norm": 1.328125,
      "learning_rate": 0.000495114932301629,
      "loss": 5.1615,
      "mean_token_accuracy": 0.19315571635961531,
      "num_tokens": 16365768.0,
      "step": 8890
    },
    {
      "entropy": 5.28054986000061,
      "epoch": 0.7429836284664216,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004951087606278328,
      "loss": 5.1238,
      "mean_token_accuracy": 0.19437289983034134,
      "num_tokens": 16375548.0,
      "step": 8895
    },
    {
      "entropy": 5.317500877380371,
      "epoch": 0.7434012696291347,
      "grad_norm": 1.125,
      "learning_rate": 0.0004951025851007678,
      "loss": 5.1975,
      "mean_token_accuracy": 0.1875303015112877,
      "num_tokens": 16385472.0,
      "step": 8900
    },
    {
      "entropy": 5.309254741668701,
      "epoch": 0.7438189107918477,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004950964057205421,
      "loss": 5.0563,
      "mean_token_accuracy": 0.19162075817584992,
      "num_tokens": 16395048.0,
      "step": 8905
    },
    {
      "entropy": 5.354975080490112,
      "epoch": 0.7442365519545606,
      "grad_norm": 1.265625,
      "learning_rate": 0.000495090222487264,
      "loss": 5.2899,
      "mean_token_accuracy": 0.18024988025426864,
      "num_tokens": 16405378.0,
      "step": 8910
    },
    {
      "entropy": 5.251638317108155,
      "epoch": 0.7446541931172737,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004950840354010415,
      "loss": 5.1645,
      "mean_token_accuracy": 0.19488231092691422,
      "num_tokens": 16415206.0,
      "step": 8915
    },
    {
      "entropy": 5.326546335220337,
      "epoch": 0.7450718342799867,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004950778444619831,
      "loss": 5.1774,
      "mean_token_accuracy": 0.19878570586442948,
      "num_tokens": 16424185.0,
      "step": 8920
    },
    {
      "entropy": 5.2738865375518795,
      "epoch": 0.7454894754426996,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000495071649670197,
      "loss": 5.1567,
      "mean_token_accuracy": 0.19145272970199584,
      "num_tokens": 16433330.0,
      "step": 8925
    },
    {
      "entropy": 5.262444305419922,
      "epoch": 0.7459071166054126,
      "grad_norm": 1.53125,
      "learning_rate": 0.000495065451025792,
      "loss": 5.1727,
      "mean_token_accuracy": 0.18978547155857087,
      "num_tokens": 16442777.0,
      "step": 8930
    },
    {
      "entropy": 5.331970071792602,
      "epoch": 0.7463247577681257,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004950592485288761,
      "loss": 5.1852,
      "mean_token_accuracy": 0.1919775828719139,
      "num_tokens": 16451838.0,
      "step": 8935
    },
    {
      "entropy": 5.254158401489258,
      "epoch": 0.7467423989308386,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004950530421795584,
      "loss": 5.1136,
      "mean_token_accuracy": 0.19258003830909728,
      "num_tokens": 16460022.0,
      "step": 8940
    },
    {
      "entropy": 5.201046943664551,
      "epoch": 0.7471600400935516,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004950468319779472,
      "loss": 5.018,
      "mean_token_accuracy": 0.20304833203554154,
      "num_tokens": 16468891.0,
      "step": 8945
    },
    {
      "entropy": 5.206154727935791,
      "epoch": 0.7475776812562647,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004950406179241515,
      "loss": 5.1149,
      "mean_token_accuracy": 0.19986711889505387,
      "num_tokens": 16478020.0,
      "step": 8950
    },
    {
      "entropy": 5.3036723136901855,
      "epoch": 0.7479953224189776,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004950344000182798,
      "loss": 5.2784,
      "mean_token_accuracy": 0.1881829619407654,
      "num_tokens": 16487536.0,
      "step": 8955
    },
    {
      "entropy": 5.335091876983642,
      "epoch": 0.7484129635816906,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004950281782604411,
      "loss": 5.1843,
      "mean_token_accuracy": 0.18343029022216797,
      "num_tokens": 16495786.0,
      "step": 8960
    },
    {
      "entropy": 5.420146846771241,
      "epoch": 0.7488306047444037,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004950219526507442,
      "loss": 5.2672,
      "mean_token_accuracy": 0.1825781837105751,
      "num_tokens": 16505535.0,
      "step": 8965
    },
    {
      "entropy": 5.388070487976075,
      "epoch": 0.7492482459071166,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004950157231892983,
      "loss": 5.2654,
      "mean_token_accuracy": 0.18759781122207642,
      "num_tokens": 16515387.0,
      "step": 8970
    },
    {
      "entropy": 5.263113594055175,
      "epoch": 0.7496658870698296,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004950094898762123,
      "loss": 5.2048,
      "mean_token_accuracy": 0.19486620426177978,
      "num_tokens": 16524421.0,
      "step": 8975
    },
    {
      "entropy": 5.432568120956421,
      "epoch": 0.7500835282325427,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004950032527115954,
      "loss": 5.2024,
      "mean_token_accuracy": 0.19057731330394745,
      "num_tokens": 16534533.0,
      "step": 8980
    },
    {
      "entropy": 5.304015398025513,
      "epoch": 0.7505011693952556,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004949970116955566,
      "loss": 5.2095,
      "mean_token_accuracy": 0.18082914799451827,
      "num_tokens": 16543712.0,
      "step": 8985
    },
    {
      "entropy": 5.2897895812988285,
      "epoch": 0.7509188105579686,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004949907668282054,
      "loss": 5.1757,
      "mean_token_accuracy": 0.19224326759576799,
      "num_tokens": 16552475.0,
      "step": 8990
    },
    {
      "entropy": 5.33959264755249,
      "epoch": 0.7513364517206816,
      "grad_norm": 1.28125,
      "learning_rate": 0.000494984518109651,
      "loss": 5.1849,
      "mean_token_accuracy": 0.18906406909227372,
      "num_tokens": 16562237.0,
      "step": 8995
    },
    {
      "entropy": 5.33980073928833,
      "epoch": 0.7517540928833946,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004949782655400028,
      "loss": 5.1846,
      "mean_token_accuracy": 0.1936530962586403,
      "num_tokens": 16570508.0,
      "step": 9000
    },
    {
      "epoch": 0.7517540928833946,
      "eval_entropy": 5.118175612241377,
      "eval_loss": 5.154400825500488,
      "eval_mean_token_accuracy": 0.19930770909194723,
      "eval_num_tokens": 16570508.0,
      "eval_runtime": 27.4929,
      "eval_samples_per_second": 1373.555,
      "eval_steps_per_second": 171.717,
      "step": 9000
    },
    {
      "entropy": 5.230000829696655,
      "epoch": 0.7521717340461076,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004949720091193704,
      "loss": 5.0694,
      "mean_token_accuracy": 0.20097985118627548,
      "num_tokens": 16578561.0,
      "step": 9005
    },
    {
      "entropy": 5.247731637954712,
      "epoch": 0.7525893752088206,
      "grad_norm": 1.28125,
      "learning_rate": 0.000494965748847863,
      "loss": 5.0536,
      "mean_token_accuracy": 0.19788717329502106,
      "num_tokens": 16587168.0,
      "step": 9010
    },
    {
      "entropy": 5.281398153305053,
      "epoch": 0.7530070163715336,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004949594847255905,
      "loss": 5.2225,
      "mean_token_accuracy": 0.18750173300504686,
      "num_tokens": 16597252.0,
      "step": 9015
    },
    {
      "entropy": 5.222692823410034,
      "epoch": 0.7534246575342466,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004949532167526625,
      "loss": 5.0651,
      "mean_token_accuracy": 0.20130424201488495,
      "num_tokens": 16607558.0,
      "step": 9020
    },
    {
      "entropy": 5.330858373641968,
      "epoch": 0.7538422986969596,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004949469449291886,
      "loss": 5.2742,
      "mean_token_accuracy": 0.17916334122419358,
      "num_tokens": 16617019.0,
      "step": 9025
    },
    {
      "entropy": 5.246083831787109,
      "epoch": 0.7542599398596725,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004949406692552787,
      "loss": 5.0962,
      "mean_token_accuracy": 0.19083139151334763,
      "num_tokens": 16625815.0,
      "step": 9030
    },
    {
      "entropy": 5.337832355499268,
      "epoch": 0.7546775810223856,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004949343897310426,
      "loss": 5.3177,
      "mean_token_accuracy": 0.18579493165016175,
      "num_tokens": 16634876.0,
      "step": 9035
    },
    {
      "entropy": 5.348042821884155,
      "epoch": 0.7550952221850986,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004949281063565904,
      "loss": 5.2071,
      "mean_token_accuracy": 0.18700264841318132,
      "num_tokens": 16643770.0,
      "step": 9040
    },
    {
      "entropy": 5.364465856552124,
      "epoch": 0.7555128633478115,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004949218191320319,
      "loss": 5.1454,
      "mean_token_accuracy": 0.19319502711296083,
      "num_tokens": 16652845.0,
      "step": 9045
    },
    {
      "entropy": 5.23198766708374,
      "epoch": 0.7559305045105246,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004949155280574772,
      "loss": 5.1129,
      "mean_token_accuracy": 0.19030310809612275,
      "num_tokens": 16662224.0,
      "step": 9050
    },
    {
      "entropy": 5.257323789596557,
      "epoch": 0.7563481456732376,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004949092331330365,
      "loss": 5.1585,
      "mean_token_accuracy": 0.1905785471200943,
      "num_tokens": 16671114.0,
      "step": 9055
    },
    {
      "entropy": 5.313417053222656,
      "epoch": 0.7567657868359505,
      "grad_norm": 1.234375,
      "learning_rate": 0.00049490293435882,
      "loss": 5.2001,
      "mean_token_accuracy": 0.17831166088581085,
      "num_tokens": 16680395.0,
      "step": 9060
    },
    {
      "entropy": 5.331158590316773,
      "epoch": 0.7571834279986636,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004948966317349379,
      "loss": 5.1553,
      "mean_token_accuracy": 0.1914367511868477,
      "num_tokens": 16690251.0,
      "step": 9065
    },
    {
      "entropy": 5.2190484523773195,
      "epoch": 0.7576010691613766,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004948903252615006,
      "loss": 5.0524,
      "mean_token_accuracy": 0.20089221596717835,
      "num_tokens": 16698945.0,
      "step": 9070
    },
    {
      "entropy": 5.346964502334595,
      "epoch": 0.7580187103240895,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004948840149386184,
      "loss": 5.2356,
      "mean_token_accuracy": 0.1882416009902954,
      "num_tokens": 16708325.0,
      "step": 9075
    },
    {
      "entropy": 5.27890191078186,
      "epoch": 0.7584363514868026,
      "grad_norm": 1.25,
      "learning_rate": 0.000494877700766402,
      "loss": 5.134,
      "mean_token_accuracy": 0.1923614338040352,
      "num_tokens": 16716392.0,
      "step": 9080
    },
    {
      "entropy": 5.322353315353394,
      "epoch": 0.7588539926495156,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004948713827449616,
      "loss": 5.1535,
      "mean_token_accuracy": 0.18847090750932693,
      "num_tokens": 16725284.0,
      "step": 9085
    },
    {
      "entropy": 5.2608133316040036,
      "epoch": 0.7592716338122285,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004948650608744081,
      "loss": 5.1832,
      "mean_token_accuracy": 0.19603832811117172,
      "num_tokens": 16734658.0,
      "step": 9090
    },
    {
      "entropy": 5.378498983383179,
      "epoch": 0.7596892749749415,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000494858735154852,
      "loss": 5.2872,
      "mean_token_accuracy": 0.18194079846143724,
      "num_tokens": 16744403.0,
      "step": 9095
    },
    {
      "entropy": 5.351605319976807,
      "epoch": 0.7601069161376546,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004948524055864041,
      "loss": 5.25,
      "mean_token_accuracy": 0.18618920296430588,
      "num_tokens": 16754254.0,
      "step": 9100
    },
    {
      "entropy": 5.324847459793091,
      "epoch": 0.7605245573003675,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004948460721691752,
      "loss": 5.2581,
      "mean_token_accuracy": 0.18689868599176407,
      "num_tokens": 16763941.0,
      "step": 9105
    },
    {
      "entropy": 5.320371913909912,
      "epoch": 0.7609421984630805,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004948397349032762,
      "loss": 5.1704,
      "mean_token_accuracy": 0.1879645839333534,
      "num_tokens": 16773181.0,
      "step": 9110
    },
    {
      "entropy": 5.185020399093628,
      "epoch": 0.7613598396257936,
      "grad_norm": 1.1875,
      "learning_rate": 0.000494833393788818,
      "loss": 5.0087,
      "mean_token_accuracy": 0.2001813605427742,
      "num_tokens": 16782622.0,
      "step": 9115
    },
    {
      "entropy": 5.317534875869751,
      "epoch": 0.7617774807885065,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004948270488259117,
      "loss": 5.2491,
      "mean_token_accuracy": 0.18950426578521729,
      "num_tokens": 16792805.0,
      "step": 9120
    },
    {
      "entropy": 5.337619018554688,
      "epoch": 0.7621951219512195,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004948207000146682,
      "loss": 5.2173,
      "mean_token_accuracy": 0.19059991836547852,
      "num_tokens": 16802641.0,
      "step": 9125
    },
    {
      "entropy": 5.402008485794068,
      "epoch": 0.7626127631139326,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004948143473551989,
      "loss": 5.1904,
      "mean_token_accuracy": 0.18393849432468415,
      "num_tokens": 16812037.0,
      "step": 9130
    },
    {
      "entropy": 5.266118907928467,
      "epoch": 0.7630304042766455,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004948079908476147,
      "loss": 5.1578,
      "mean_token_accuracy": 0.20224610567092896,
      "num_tokens": 16820279.0,
      "step": 9135
    },
    {
      "entropy": 5.282298469543457,
      "epoch": 0.7634480454393585,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004948016304920271,
      "loss": 5.1504,
      "mean_token_accuracy": 0.1833828017115593,
      "num_tokens": 16829396.0,
      "step": 9140
    },
    {
      "entropy": 5.291776180267334,
      "epoch": 0.7638656866020715,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004947952662885473,
      "loss": 5.1396,
      "mean_token_accuracy": 0.19542227536439896,
      "num_tokens": 16838389.0,
      "step": 9145
    },
    {
      "entropy": 5.3618017673492435,
      "epoch": 0.7642833277647845,
      "grad_norm": 1.28125,
      "learning_rate": 0.000494788898237287,
      "loss": 5.2449,
      "mean_token_accuracy": 0.1881488025188446,
      "num_tokens": 16848012.0,
      "step": 9150
    },
    {
      "entropy": 5.366077709197998,
      "epoch": 0.7647009689274975,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004947825263383573,
      "loss": 5.205,
      "mean_token_accuracy": 0.1885187953710556,
      "num_tokens": 16857591.0,
      "step": 9155
    },
    {
      "entropy": 5.3177809715271,
      "epoch": 0.7651186100902105,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00049477615059187,
      "loss": 5.2032,
      "mean_token_accuracy": 0.18001774251461028,
      "num_tokens": 16866792.0,
      "step": 9160
    },
    {
      "entropy": 5.252402687072754,
      "epoch": 0.7655362512529235,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004947697709979367,
      "loss": 5.2147,
      "mean_token_accuracy": 0.18903812170028686,
      "num_tokens": 16876602.0,
      "step": 9165
    },
    {
      "entropy": 5.315468978881836,
      "epoch": 0.7659538924156365,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004947633875566688,
      "loss": 5.2035,
      "mean_token_accuracy": 0.18632730841636658,
      "num_tokens": 16886992.0,
      "step": 9170
    },
    {
      "entropy": 5.304830694198609,
      "epoch": 0.7663715335783495,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004947570002681785,
      "loss": 5.1325,
      "mean_token_accuracy": 0.19268832355737686,
      "num_tokens": 16895716.0,
      "step": 9175
    },
    {
      "entropy": 5.290454626083374,
      "epoch": 0.7667891747410625,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004947506091325772,
      "loss": 5.1253,
      "mean_token_accuracy": 0.20394227504730225,
      "num_tokens": 16904568.0,
      "step": 9180
    },
    {
      "entropy": 5.3105532169342045,
      "epoch": 0.7672068159037755,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004947442141499771,
      "loss": 5.2281,
      "mean_token_accuracy": 0.18760323226451875,
      "num_tokens": 16913667.0,
      "step": 9185
    },
    {
      "entropy": 5.29073166847229,
      "epoch": 0.7676244570664885,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004947378153204899,
      "loss": 5.1561,
      "mean_token_accuracy": 0.18824365586042405,
      "num_tokens": 16922897.0,
      "step": 9190
    },
    {
      "entropy": 5.369127941131592,
      "epoch": 0.7680420982292014,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004947314126442279,
      "loss": 5.1741,
      "mean_token_accuracy": 0.18450040221214295,
      "num_tokens": 16931559.0,
      "step": 9195
    },
    {
      "entropy": 5.281914186477661,
      "epoch": 0.7684597393919145,
      "grad_norm": 1.453125,
      "learning_rate": 0.000494725006121303,
      "loss": 5.1677,
      "mean_token_accuracy": 0.18040622174739837,
      "num_tokens": 16940191.0,
      "step": 9200
    },
    {
      "entropy": 5.311995553970337,
      "epoch": 0.7688773805546275,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004947185957518273,
      "loss": 5.2192,
      "mean_token_accuracy": 0.1866691917181015,
      "num_tokens": 16948971.0,
      "step": 9205
    },
    {
      "entropy": 5.3018656253814695,
      "epoch": 0.7692950217173404,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004947121815359133,
      "loss": 5.1516,
      "mean_token_accuracy": 0.1922988548874855,
      "num_tokens": 16957944.0,
      "step": 9210
    },
    {
      "entropy": 5.231423616409302,
      "epoch": 0.7697126628800535,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004947057634736729,
      "loss": 5.1336,
      "mean_token_accuracy": 0.1856879249215126,
      "num_tokens": 16967418.0,
      "step": 9215
    },
    {
      "entropy": 5.231414794921875,
      "epoch": 0.7701303040427665,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004946993415652186,
      "loss": 5.1256,
      "mean_token_accuracy": 0.19581584632396698,
      "num_tokens": 16977233.0,
      "step": 9220
    },
    {
      "entropy": 5.248545694351196,
      "epoch": 0.7705479452054794,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000494692915810663,
      "loss": 5.1744,
      "mean_token_accuracy": 0.1953287437558174,
      "num_tokens": 16986023.0,
      "step": 9225
    },
    {
      "entropy": 5.286213159561157,
      "epoch": 0.7709655863681925,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004946864862101185,
      "loss": 5.1198,
      "mean_token_accuracy": 0.1882033348083496,
      "num_tokens": 16994987.0,
      "step": 9230
    },
    {
      "entropy": 5.235636520385742,
      "epoch": 0.7713832275309055,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004946800527636975,
      "loss": 5.1351,
      "mean_token_accuracy": 0.18983470499515534,
      "num_tokens": 17003902.0,
      "step": 9235
    },
    {
      "entropy": 5.340488433837891,
      "epoch": 0.7718008686936184,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004946736154715129,
      "loss": 5.2271,
      "mean_token_accuracy": 0.18435494005680084,
      "num_tokens": 17013258.0,
      "step": 9240
    },
    {
      "entropy": 5.379638051986694,
      "epoch": 0.7722185098563314,
      "grad_norm": 1.25,
      "learning_rate": 0.0004946671743336771,
      "loss": 5.225,
      "mean_token_accuracy": 0.18321820199489594,
      "num_tokens": 17022258.0,
      "step": 9245
    },
    {
      "entropy": 5.296166706085205,
      "epoch": 0.7726361510190445,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004946607293503031,
      "loss": 5.1419,
      "mean_token_accuracy": 0.18779005855321884,
      "num_tokens": 17031485.0,
      "step": 9250
    },
    {
      "entropy": 5.161727809906006,
      "epoch": 0.7730537921817574,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004946542805215036,
      "loss": 5.105,
      "mean_token_accuracy": 0.1998660609126091,
      "num_tokens": 17039916.0,
      "step": 9255
    },
    {
      "entropy": 5.204945659637451,
      "epoch": 0.7734714333444704,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004946478278473914,
      "loss": 5.1482,
      "mean_token_accuracy": 0.1925191819667816,
      "num_tokens": 17048153.0,
      "step": 9260
    },
    {
      "entropy": 5.283808374404908,
      "epoch": 0.7738890745071835,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004946413713280797,
      "loss": 5.1902,
      "mean_token_accuracy": 0.1889582023024559,
      "num_tokens": 17057473.0,
      "step": 9265
    },
    {
      "entropy": 5.319660758972168,
      "epoch": 0.7743067156698964,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004946349109636814,
      "loss": 5.1876,
      "mean_token_accuracy": 0.19380863904953002,
      "num_tokens": 17066431.0,
      "step": 9270
    },
    {
      "entropy": 5.343507146835327,
      "epoch": 0.7747243568326094,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004946284467543096,
      "loss": 5.2842,
      "mean_token_accuracy": 0.18179262429475784,
      "num_tokens": 17076519.0,
      "step": 9275
    },
    {
      "entropy": 5.33641676902771,
      "epoch": 0.7751419979953225,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004946219787000776,
      "loss": 5.1992,
      "mean_token_accuracy": 0.19015321731567383,
      "num_tokens": 17084651.0,
      "step": 9280
    },
    {
      "entropy": 5.377818202972412,
      "epoch": 0.7755596391580354,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004946155068010984,
      "loss": 5.2523,
      "mean_token_accuracy": 0.1837206155061722,
      "num_tokens": 17094943.0,
      "step": 9285
    },
    {
      "entropy": 5.260235357284546,
      "epoch": 0.7759772803207484,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004946090310574854,
      "loss": 5.1129,
      "mean_token_accuracy": 0.19882804453372954,
      "num_tokens": 17103548.0,
      "step": 9290
    },
    {
      "entropy": 5.2598082542419435,
      "epoch": 0.7763949214834615,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004946025514693519,
      "loss": 5.0896,
      "mean_token_accuracy": 0.18809034079313278,
      "num_tokens": 17112472.0,
      "step": 9295
    },
    {
      "entropy": 5.397208881378174,
      "epoch": 0.7768125626461744,
      "grad_norm": 1.375,
      "learning_rate": 0.0004945960680368115,
      "loss": 5.2297,
      "mean_token_accuracy": 0.1803405612707138,
      "num_tokens": 17121536.0,
      "step": 9300
    },
    {
      "entropy": 5.406971025466919,
      "epoch": 0.7772302038088874,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004945895807599775,
      "loss": 5.2584,
      "mean_token_accuracy": 0.1797230288386345,
      "num_tokens": 17130708.0,
      "step": 9305
    },
    {
      "entropy": 5.200753116607666,
      "epoch": 0.7776478449716004,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004945830896389636,
      "loss": 5.0601,
      "mean_token_accuracy": 0.1990275502204895,
      "num_tokens": 17139437.0,
      "step": 9310
    },
    {
      "entropy": 5.261353635787964,
      "epoch": 0.7780654861343134,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004945765946738835,
      "loss": 5.2228,
      "mean_token_accuracy": 0.18262001425027846,
      "num_tokens": 17149107.0,
      "step": 9315
    },
    {
      "entropy": 5.334297466278076,
      "epoch": 0.7784831272970264,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004945700958648507,
      "loss": 5.2099,
      "mean_token_accuracy": 0.18105990290641785,
      "num_tokens": 17158866.0,
      "step": 9320
    },
    {
      "entropy": 5.318431806564331,
      "epoch": 0.7789007684597394,
      "grad_norm": 1.375,
      "learning_rate": 0.000494563593211979,
      "loss": 5.0895,
      "mean_token_accuracy": 0.19431317895650863,
      "num_tokens": 17167909.0,
      "step": 9325
    },
    {
      "entropy": 5.243116664886474,
      "epoch": 0.7793184096224524,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004945570867153826,
      "loss": 5.2008,
      "mean_token_accuracy": 0.18520733267068862,
      "num_tokens": 17176746.0,
      "step": 9330
    },
    {
      "entropy": 5.293108367919922,
      "epoch": 0.7797360507851654,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004945505763751749,
      "loss": 5.2054,
      "mean_token_accuracy": 0.18373152613639832,
      "num_tokens": 17186219.0,
      "step": 9335
    },
    {
      "entropy": 5.311901235580445,
      "epoch": 0.7801536919478784,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004945440621914702,
      "loss": 5.0486,
      "mean_token_accuracy": 0.1980719268321991,
      "num_tokens": 17195811.0,
      "step": 9340
    },
    {
      "entropy": 5.20134711265564,
      "epoch": 0.7805713331105913,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004945375441643823,
      "loss": 5.2104,
      "mean_token_accuracy": 0.19056608378887177,
      "num_tokens": 17205333.0,
      "step": 9345
    },
    {
      "entropy": 5.32084846496582,
      "epoch": 0.7809889742733044,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004945310222940254,
      "loss": 5.1458,
      "mean_token_accuracy": 0.186599463224411,
      "num_tokens": 17214493.0,
      "step": 9350
    },
    {
      "entropy": 5.373107767105102,
      "epoch": 0.7814066154360174,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004945244965805138,
      "loss": 5.1928,
      "mean_token_accuracy": 0.18953339904546737,
      "num_tokens": 17223602.0,
      "step": 9355
    },
    {
      "entropy": 5.3266363620758055,
      "epoch": 0.7818242565987303,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004945179670239616,
      "loss": 5.212,
      "mean_token_accuracy": 0.1905250996351242,
      "num_tokens": 17232915.0,
      "step": 9360
    },
    {
      "entropy": 5.307262325286866,
      "epoch": 0.7822418977614434,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004945114336244832,
      "loss": 5.1893,
      "mean_token_accuracy": 0.18642881214618684,
      "num_tokens": 17242489.0,
      "step": 9365
    },
    {
      "entropy": 5.167524290084839,
      "epoch": 0.7826595389241564,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000494504896382193,
      "loss": 4.947,
      "mean_token_accuracy": 0.20279575288295745,
      "num_tokens": 17251496.0,
      "step": 9370
    },
    {
      "entropy": 5.219880247116089,
      "epoch": 0.7830771800868693,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004944983552972053,
      "loss": 5.1339,
      "mean_token_accuracy": 0.1955128401517868,
      "num_tokens": 17260550.0,
      "step": 9375
    },
    {
      "entropy": 5.286794233322143,
      "epoch": 0.7834948212495824,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004944918103696347,
      "loss": 5.0946,
      "mean_token_accuracy": 0.19075968712568284,
      "num_tokens": 17269085.0,
      "step": 9380
    },
    {
      "entropy": 5.394890975952149,
      "epoch": 0.7839124624122954,
      "grad_norm": 1.25,
      "learning_rate": 0.0004944852615995956,
      "loss": 5.3224,
      "mean_token_accuracy": 0.1880297228693962,
      "num_tokens": 17278766.0,
      "step": 9385
    },
    {
      "entropy": 5.433079814910888,
      "epoch": 0.7843301035750083,
      "grad_norm": 1.40625,
      "learning_rate": 0.000494478708987203,
      "loss": 5.2053,
      "mean_token_accuracy": 0.18566980659961702,
      "num_tokens": 17288316.0,
      "step": 9390
    },
    {
      "entropy": 5.218790531158447,
      "epoch": 0.7847477447377214,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004944721525325713,
      "loss": 5.0423,
      "mean_token_accuracy": 0.19153362065553664,
      "num_tokens": 17297829.0,
      "step": 9395
    },
    {
      "entropy": 5.236908483505249,
      "epoch": 0.7851653859004344,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004944655922358154,
      "loss": 5.1301,
      "mean_token_accuracy": 0.18671831637620925,
      "num_tokens": 17306631.0,
      "step": 9400
    },
    {
      "entropy": 5.246349096298218,
      "epoch": 0.7855830270631473,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004944590280970503,
      "loss": 5.1225,
      "mean_token_accuracy": 0.19494060426950455,
      "num_tokens": 17316829.0,
      "step": 9405
    },
    {
      "entropy": 5.297109460830688,
      "epoch": 0.7860006682258603,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004944524601163905,
      "loss": 5.1656,
      "mean_token_accuracy": 0.1853885069489479,
      "num_tokens": 17326568.0,
      "step": 9410
    },
    {
      "entropy": 5.321149778366089,
      "epoch": 0.7864183093885734,
      "grad_norm": 1.375,
      "learning_rate": 0.0004944458882939513,
      "loss": 5.2286,
      "mean_token_accuracy": 0.1837119311094284,
      "num_tokens": 17335524.0,
      "step": 9415
    },
    {
      "entropy": 5.358046627044677,
      "epoch": 0.7868359505512863,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004944393126298478,
      "loss": 5.2465,
      "mean_token_accuracy": 0.18291445523500444,
      "num_tokens": 17346198.0,
      "step": 9420
    },
    {
      "entropy": 5.287853765487671,
      "epoch": 0.7872535917139993,
      "grad_norm": 1.25,
      "learning_rate": 0.000494432733124195,
      "loss": 5.0982,
      "mean_token_accuracy": 0.19188064485788345,
      "num_tokens": 17354696.0,
      "step": 9425
    },
    {
      "entropy": 5.201977014541626,
      "epoch": 0.7876712328767124,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000494426149777108,
      "loss": 5.0776,
      "mean_token_accuracy": 0.19119950830936433,
      "num_tokens": 17363314.0,
      "step": 9430
    },
    {
      "entropy": 5.219988584518433,
      "epoch": 0.7880888740394253,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000494419562588702,
      "loss": 5.1357,
      "mean_token_accuracy": 0.19479642510414125,
      "num_tokens": 17372662.0,
      "step": 9435
    },
    {
      "entropy": 5.214144897460938,
      "epoch": 0.7885065152021383,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004944129715590927,
      "loss": 5.1034,
      "mean_token_accuracy": 0.19101351946592332,
      "num_tokens": 17381931.0,
      "step": 9440
    },
    {
      "entropy": 5.357375192642212,
      "epoch": 0.7889241563648514,
      "grad_norm": 1.21875,
      "learning_rate": 0.000494406376688395,
      "loss": 5.1995,
      "mean_token_accuracy": 0.18180252313613893,
      "num_tokens": 17391210.0,
      "step": 9445
    },
    {
      "entropy": 5.222433948516846,
      "epoch": 0.7893417975275643,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004943997779767248,
      "loss": 5.0655,
      "mean_token_accuracy": 0.2050221011042595,
      "num_tokens": 17401102.0,
      "step": 9450
    },
    {
      "entropy": 5.204224348068237,
      "epoch": 0.7897594386902773,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004943931754241972,
      "loss": 5.1087,
      "mean_token_accuracy": 0.19259266704320907,
      "num_tokens": 17410036.0,
      "step": 9455
    },
    {
      "entropy": 5.275397205352784,
      "epoch": 0.7901770798529903,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000494386569030928,
      "loss": 5.1621,
      "mean_token_accuracy": 0.18584441691637038,
      "num_tokens": 17419348.0,
      "step": 9460
    },
    {
      "entropy": 5.324286556243896,
      "epoch": 0.7905947210157033,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004943799587970328,
      "loss": 5.1578,
      "mean_token_accuracy": 0.18599013537168502,
      "num_tokens": 17427937.0,
      "step": 9465
    },
    {
      "entropy": 5.178501033782959,
      "epoch": 0.7910123621784163,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004943733447226275,
      "loss": 5.0317,
      "mean_token_accuracy": 0.19731151312589645,
      "num_tokens": 17436259.0,
      "step": 9470
    },
    {
      "entropy": 5.338555812835693,
      "epoch": 0.7914300033411293,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004943667268078277,
      "loss": 5.2408,
      "mean_token_accuracy": 0.18015268296003342,
      "num_tokens": 17445202.0,
      "step": 9475
    },
    {
      "entropy": 5.196104383468628,
      "epoch": 0.7918476445038423,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004943601050527493,
      "loss": 5.0995,
      "mean_token_accuracy": 0.19210998266935347,
      "num_tokens": 17453928.0,
      "step": 9480
    },
    {
      "entropy": 5.265049266815185,
      "epoch": 0.7922652856665553,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004943534794575081,
      "loss": 5.1487,
      "mean_token_accuracy": 0.19385261386632918,
      "num_tokens": 17462609.0,
      "step": 9485
    },
    {
      "entropy": 5.348389196395874,
      "epoch": 0.7926829268292683,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004943468500222203,
      "loss": 5.1729,
      "mean_token_accuracy": 0.19015553295612336,
      "num_tokens": 17471326.0,
      "step": 9490
    },
    {
      "entropy": 5.1824195861816404,
      "epoch": 0.7931005679919813,
      "grad_norm": 1.25,
      "learning_rate": 0.0004943402167470019,
      "loss": 5.1353,
      "mean_token_accuracy": 0.1968790963292122,
      "num_tokens": 17480296.0,
      "step": 9495
    },
    {
      "entropy": 5.1844724178314205,
      "epoch": 0.7935182091546943,
      "grad_norm": 1.3125,
      "learning_rate": 0.000494333579631969,
      "loss": 5.0726,
      "mean_token_accuracy": 0.1936543330550194,
      "num_tokens": 17489779.0,
      "step": 9500
    },
    {
      "entropy": 5.295859098434448,
      "epoch": 0.7939358503174073,
      "grad_norm": 1.25,
      "learning_rate": 0.0004943269386772376,
      "loss": 5.1725,
      "mean_token_accuracy": 0.18356461971998214,
      "num_tokens": 17499796.0,
      "step": 9505
    },
    {
      "entropy": 5.322860479354858,
      "epoch": 0.7943534914801202,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004943202938829243,
      "loss": 5.2631,
      "mean_token_accuracy": 0.18240389674901963,
      "num_tokens": 17509223.0,
      "step": 9510
    },
    {
      "entropy": 5.297752571105957,
      "epoch": 0.7947711326428333,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004943136452491452,
      "loss": 5.17,
      "mean_token_accuracy": 0.1856241211295128,
      "num_tokens": 17518712.0,
      "step": 9515
    },
    {
      "entropy": 5.205142688751221,
      "epoch": 0.7951887738055463,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004943069927760168,
      "loss": 5.1045,
      "mean_token_accuracy": 0.1932361900806427,
      "num_tokens": 17528648.0,
      "step": 9520
    },
    {
      "entropy": 5.133691883087158,
      "epoch": 0.7956064149682592,
      "grad_norm": 1.25,
      "learning_rate": 0.0004943003364636554,
      "loss": 5.1179,
      "mean_token_accuracy": 0.18653893023729323,
      "num_tokens": 17537859.0,
      "step": 9525
    },
    {
      "entropy": 5.351423740386963,
      "epoch": 0.7960240561309723,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004942936763121777,
      "loss": 5.2608,
      "mean_token_accuracy": 0.18303751498460769,
      "num_tokens": 17546438.0,
      "step": 9530
    },
    {
      "entropy": 5.338749694824219,
      "epoch": 0.7964416972936853,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004942870123217002,
      "loss": 5.0919,
      "mean_token_accuracy": 0.1958828955888748,
      "num_tokens": 17554537.0,
      "step": 9535
    },
    {
      "entropy": 5.379541683197021,
      "epoch": 0.7968593384563982,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004942803444923398,
      "loss": 5.3016,
      "mean_token_accuracy": 0.18473671078681947,
      "num_tokens": 17564239.0,
      "step": 9540
    },
    {
      "entropy": 5.263526868820191,
      "epoch": 0.7972769796191113,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004942736728242127,
      "loss": 5.1708,
      "mean_token_accuracy": 0.1941849857568741,
      "num_tokens": 17574437.0,
      "step": 9545
    },
    {
      "entropy": 5.177925062179566,
      "epoch": 0.7976946207818243,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004942669973174363,
      "loss": 5.1119,
      "mean_token_accuracy": 0.19489021897315978,
      "num_tokens": 17583173.0,
      "step": 9550
    },
    {
      "entropy": 5.26393346786499,
      "epoch": 0.7981122619445372,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000494260317972127,
      "loss": 5.1687,
      "mean_token_accuracy": 0.1888450115919113,
      "num_tokens": 17591731.0,
      "step": 9555
    },
    {
      "entropy": 5.335217618942261,
      "epoch": 0.7985299031072502,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004942536347884019,
      "loss": 5.1747,
      "mean_token_accuracy": 0.18973340094089508,
      "num_tokens": 17600778.0,
      "step": 9560
    },
    {
      "entropy": 5.235171461105347,
      "epoch": 0.7989475442699633,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004942469477663782,
      "loss": 5.15,
      "mean_token_accuracy": 0.2003604143857956,
      "num_tokens": 17609470.0,
      "step": 9565
    },
    {
      "entropy": 5.18988299369812,
      "epoch": 0.7993651854326762,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004942402569061726,
      "loss": 5.0359,
      "mean_token_accuracy": 0.19401865899562837,
      "num_tokens": 17618434.0,
      "step": 9570
    },
    {
      "entropy": 5.215825700759888,
      "epoch": 0.7997828265953892,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004942335622079025,
      "loss": 5.1706,
      "mean_token_accuracy": 0.18902689218521118,
      "num_tokens": 17628170.0,
      "step": 9575
    },
    {
      "entropy": 5.22637243270874,
      "epoch": 0.8002004677581023,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004942268636716848,
      "loss": 5.1036,
      "mean_token_accuracy": 0.19581829905509948,
      "num_tokens": 17637391.0,
      "step": 9580
    },
    {
      "entropy": 5.286163759231568,
      "epoch": 0.8006181089208152,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004942201612976372,
      "loss": 5.1805,
      "mean_token_accuracy": 0.19466704428195952,
      "num_tokens": 17647466.0,
      "step": 9585
    },
    {
      "entropy": 5.2615069389343265,
      "epoch": 0.8010357500835282,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004942134550858767,
      "loss": 5.103,
      "mean_token_accuracy": 0.19750634878873824,
      "num_tokens": 17656155.0,
      "step": 9590
    },
    {
      "entropy": 5.346674251556396,
      "epoch": 0.8014533912462413,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004942067450365208,
      "loss": 5.2125,
      "mean_token_accuracy": 0.19068874418735504,
      "num_tokens": 17665242.0,
      "step": 9595
    },
    {
      "entropy": 5.287128686904907,
      "epoch": 0.8018710324089542,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004942000311496871,
      "loss": 5.2343,
      "mean_token_accuracy": 0.1857941746711731,
      "num_tokens": 17674964.0,
      "step": 9600
    },
    {
      "entropy": 5.340595245361328,
      "epoch": 0.8022886735716672,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004941933134254928,
      "loss": 5.2241,
      "mean_token_accuracy": 0.1916162669658661,
      "num_tokens": 17683275.0,
      "step": 9605
    },
    {
      "entropy": 5.220560932159424,
      "epoch": 0.8027063147343803,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004941865918640559,
      "loss": 5.1659,
      "mean_token_accuracy": 0.18791708052158357,
      "num_tokens": 17691841.0,
      "step": 9610
    },
    {
      "entropy": 5.292420291900635,
      "epoch": 0.8031239558970932,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004941798664654937,
      "loss": 5.1364,
      "mean_token_accuracy": 0.19378622025251388,
      "num_tokens": 17701258.0,
      "step": 9615
    },
    {
      "entropy": 5.3023679733276365,
      "epoch": 0.8035415970598062,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004941731372299241,
      "loss": 5.228,
      "mean_token_accuracy": 0.18204617351293564,
      "num_tokens": 17711028.0,
      "step": 9620
    },
    {
      "entropy": 5.27116641998291,
      "epoch": 0.8039592382225192,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004941664041574651,
      "loss": 5.1567,
      "mean_token_accuracy": 0.20359378606081008,
      "num_tokens": 17720175.0,
      "step": 9625
    },
    {
      "entropy": 5.291797780990601,
      "epoch": 0.8043768793852322,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004941596672482343,
      "loss": 5.1266,
      "mean_token_accuracy": 0.19067127704620362,
      "num_tokens": 17730420.0,
      "step": 9630
    },
    {
      "entropy": 5.332720518112183,
      "epoch": 0.8047945205479452,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004941529265023498,
      "loss": 5.2326,
      "mean_token_accuracy": 0.18311050534248352,
      "num_tokens": 17739802.0,
      "step": 9635
    },
    {
      "entropy": 5.33508825302124,
      "epoch": 0.8052121617106582,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004941461819199293,
      "loss": 5.1207,
      "mean_token_accuracy": 0.19688181728124618,
      "num_tokens": 17749450.0,
      "step": 9640
    },
    {
      "entropy": 5.335902452468872,
      "epoch": 0.8056298028733712,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004941394335010912,
      "loss": 5.3166,
      "mean_token_accuracy": 0.1839230477809906,
      "num_tokens": 17758733.0,
      "step": 9645
    },
    {
      "entropy": 5.160353755950927,
      "epoch": 0.8060474440360842,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004941326812459535,
      "loss": 5.2163,
      "mean_token_accuracy": 0.18263708800077438,
      "num_tokens": 17769037.0,
      "step": 9650
    },
    {
      "entropy": 5.276035165786743,
      "epoch": 0.8064650851987972,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004941259251546346,
      "loss": 5.1491,
      "mean_token_accuracy": 0.186018468439579,
      "num_tokens": 17777603.0,
      "step": 9655
    },
    {
      "entropy": 5.277223825454712,
      "epoch": 0.8068827263615101,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004941191652272526,
      "loss": 5.1573,
      "mean_token_accuracy": 0.18110825270414352,
      "num_tokens": 17787513.0,
      "step": 9660
    },
    {
      "entropy": 5.20878586769104,
      "epoch": 0.8073003675242232,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004941124014639258,
      "loss": 5.1502,
      "mean_token_accuracy": 0.1897517502307892,
      "num_tokens": 17796720.0,
      "step": 9665
    },
    {
      "entropy": 5.299870395660401,
      "epoch": 0.8077180086869362,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004941056338647726,
      "loss": 5.1315,
      "mean_token_accuracy": 0.19099827855825424,
      "num_tokens": 17806024.0,
      "step": 9670
    },
    {
      "entropy": 5.239360809326172,
      "epoch": 0.8081356498496491,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004940988624299116,
      "loss": 5.0881,
      "mean_token_accuracy": 0.19406146854162215,
      "num_tokens": 17815247.0,
      "step": 9675
    },
    {
      "entropy": 5.361795806884766,
      "epoch": 0.8085532910123622,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004940920871594613,
      "loss": 5.1534,
      "mean_token_accuracy": 0.19756268560886384,
      "num_tokens": 17824024.0,
      "step": 9680
    },
    {
      "entropy": 5.221467924118042,
      "epoch": 0.8089709321750752,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004940853080535403,
      "loss": 5.1128,
      "mean_token_accuracy": 0.18864771723747253,
      "num_tokens": 17832782.0,
      "step": 9685
    },
    {
      "entropy": 5.22515721321106,
      "epoch": 0.8093885733377881,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004940785251122673,
      "loss": 5.1082,
      "mean_token_accuracy": 0.19930712431669234,
      "num_tokens": 17841961.0,
      "step": 9690
    },
    {
      "entropy": 5.264921712875366,
      "epoch": 0.8098062145005012,
      "grad_norm": 1.25,
      "learning_rate": 0.0004940717383357608,
      "loss": 5.2102,
      "mean_token_accuracy": 0.19785915017127992,
      "num_tokens": 17851364.0,
      "step": 9695
    },
    {
      "entropy": 5.411531400680542,
      "epoch": 0.8102238556632142,
      "grad_norm": 1.25,
      "learning_rate": 0.0004940649477241401,
      "loss": 5.2693,
      "mean_token_accuracy": 0.17801151424646378,
      "num_tokens": 17860871.0,
      "step": 9700
    },
    {
      "entropy": 5.280926990509033,
      "epoch": 0.8106414968259271,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004940581532775237,
      "loss": 5.1823,
      "mean_token_accuracy": 0.19045751094818114,
      "num_tokens": 17869687.0,
      "step": 9705
    },
    {
      "entropy": 5.2791155815124515,
      "epoch": 0.8110591379886402,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004940513549960306,
      "loss": 5.1234,
      "mean_token_accuracy": 0.1878678858280182,
      "num_tokens": 17878815.0,
      "step": 9710
    },
    {
      "entropy": 5.243163347244263,
      "epoch": 0.8114767791513532,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004940445528797798,
      "loss": 5.0625,
      "mean_token_accuracy": 0.19183314144611358,
      "num_tokens": 17888302.0,
      "step": 9715
    },
    {
      "entropy": 5.162750244140625,
      "epoch": 0.8118944203140661,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004940377469288906,
      "loss": 5.0381,
      "mean_token_accuracy": 0.1982324853539467,
      "num_tokens": 17896807.0,
      "step": 9720
    },
    {
      "entropy": 5.217282009124756,
      "epoch": 0.8123120614767791,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004940309371434819,
      "loss": 5.0724,
      "mean_token_accuracy": 0.19751788824796676,
      "num_tokens": 17905988.0,
      "step": 9725
    },
    {
      "entropy": 5.325047874450684,
      "epoch": 0.8127297026394922,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004940241235236731,
      "loss": 5.2234,
      "mean_token_accuracy": 0.18605109602212905,
      "num_tokens": 17915917.0,
      "step": 9730
    },
    {
      "entropy": 5.322859477996826,
      "epoch": 0.8131473438022051,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004940173060695833,
      "loss": 5.2218,
      "mean_token_accuracy": 0.18650235384702682,
      "num_tokens": 17926158.0,
      "step": 9735
    },
    {
      "entropy": 5.285133743286133,
      "epoch": 0.8135649849649181,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004940104847813319,
      "loss": 5.1694,
      "mean_token_accuracy": 0.1920665755867958,
      "num_tokens": 17935488.0,
      "step": 9740
    },
    {
      "entropy": 5.338223314285278,
      "epoch": 0.8139826261276312,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004940036596590384,
      "loss": 5.087,
      "mean_token_accuracy": 0.1884175270795822,
      "num_tokens": 17944850.0,
      "step": 9745
    },
    {
      "entropy": 5.244886875152588,
      "epoch": 0.8144002672903441,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004939968307028221,
      "loss": 5.1374,
      "mean_token_accuracy": 0.18965592384338378,
      "num_tokens": 17954796.0,
      "step": 9750
    },
    {
      "entropy": 5.187929439544678,
      "epoch": 0.8148179084530571,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004939899979128028,
      "loss": 5.1138,
      "mean_token_accuracy": 0.19805856496095658,
      "num_tokens": 17963875.0,
      "step": 9755
    },
    {
      "entropy": 5.252599382400513,
      "epoch": 0.8152355496157702,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004939831612890999,
      "loss": 5.1731,
      "mean_token_accuracy": 0.1915595605969429,
      "num_tokens": 17972789.0,
      "step": 9760
    },
    {
      "entropy": 5.211603832244873,
      "epoch": 0.8156531907784831,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004939763208318332,
      "loss": 5.1544,
      "mean_token_accuracy": 0.19271610230207442,
      "num_tokens": 17982355.0,
      "step": 9765
    },
    {
      "entropy": 5.360373783111572,
      "epoch": 0.8160708319411961,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004939694765411225,
      "loss": 5.1854,
      "mean_token_accuracy": 0.18297222703695298,
      "num_tokens": 17992046.0,
      "step": 9770
    },
    {
      "entropy": 5.227514982223511,
      "epoch": 0.8164884731039092,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004939626284170874,
      "loss": 5.1061,
      "mean_token_accuracy": 0.20245727598667146,
      "num_tokens": 18001440.0,
      "step": 9775
    },
    {
      "entropy": 5.355906438827515,
      "epoch": 0.8169061142666221,
      "grad_norm": 1.4609375,
      "learning_rate": 0.000493955776459848,
      "loss": 5.2502,
      "mean_token_accuracy": 0.18334511071443557,
      "num_tokens": 18010817.0,
      "step": 9780
    },
    {
      "entropy": 5.308782291412354,
      "epoch": 0.8173237554293351,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004939489206695241,
      "loss": 5.1645,
      "mean_token_accuracy": 0.1904589980840683,
      "num_tokens": 18019547.0,
      "step": 9785
    },
    {
      "entropy": 5.229141664505005,
      "epoch": 0.8177413965920481,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004939420610462359,
      "loss": 5.18,
      "mean_token_accuracy": 0.18796873241662979,
      "num_tokens": 18029191.0,
      "step": 9790
    },
    {
      "entropy": 5.345380783081055,
      "epoch": 0.8181590377547611,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004939351975901034,
      "loss": 5.2233,
      "mean_token_accuracy": 0.1831236332654953,
      "num_tokens": 18038831.0,
      "step": 9795
    },
    {
      "entropy": 5.295958232879639,
      "epoch": 0.8185766789174741,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004939283303012467,
      "loss": 5.1313,
      "mean_token_accuracy": 0.19052996933460237,
      "num_tokens": 18049243.0,
      "step": 9800
    },
    {
      "entropy": 5.170895385742187,
      "epoch": 0.8189943200801871,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000493921459179786,
      "loss": 5.0891,
      "mean_token_accuracy": 0.19231677800416946,
      "num_tokens": 18057960.0,
      "step": 9805
    },
    {
      "entropy": 5.316111946105957,
      "epoch": 0.8194119612429,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004939145842258418,
      "loss": 5.2026,
      "mean_token_accuracy": 0.18537302017211915,
      "num_tokens": 18066677.0,
      "step": 9810
    },
    {
      "entropy": 5.342463731765747,
      "epoch": 0.8198296024056131,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004939077054395341,
      "loss": 5.1626,
      "mean_token_accuracy": 0.1867675796151161,
      "num_tokens": 18075830.0,
      "step": 9815
    },
    {
      "entropy": 5.26358733177185,
      "epoch": 0.8202472435683261,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004939008228209836,
      "loss": 5.0732,
      "mean_token_accuracy": 0.18784022331237793,
      "num_tokens": 18085335.0,
      "step": 9820
    },
    {
      "entropy": 5.280162191390991,
      "epoch": 0.820664884731039,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004938939363703106,
      "loss": 5.1404,
      "mean_token_accuracy": 0.18583147525787352,
      "num_tokens": 18094531.0,
      "step": 9825
    },
    {
      "entropy": 5.246564626693726,
      "epoch": 0.8210825258937521,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004938870460876359,
      "loss": 5.1098,
      "mean_token_accuracy": 0.19355699717998504,
      "num_tokens": 18103593.0,
      "step": 9830
    },
    {
      "entropy": 5.298299646377563,
      "epoch": 0.8215001670564651,
      "grad_norm": 1.265625,
      "learning_rate": 0.00049388015197308,
      "loss": 5.1697,
      "mean_token_accuracy": 0.18958255052566528,
      "num_tokens": 18112826.0,
      "step": 9835
    },
    {
      "entropy": 5.29297833442688,
      "epoch": 0.821917808219178,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004938732540267635,
      "loss": 5.2317,
      "mean_token_accuracy": 0.1903190493583679,
      "num_tokens": 18121687.0,
      "step": 9840
    },
    {
      "entropy": 5.302140951156616,
      "epoch": 0.8223354493818911,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004938663522488072,
      "loss": 5.101,
      "mean_token_accuracy": 0.19614289999008178,
      "num_tokens": 18130071.0,
      "step": 9845
    },
    {
      "entropy": 5.364624977111816,
      "epoch": 0.8227530905446041,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004938594466393321,
      "loss": 5.3051,
      "mean_token_accuracy": 0.18107421696186066,
      "num_tokens": 18139077.0,
      "step": 9850
    },
    {
      "entropy": 5.294253635406494,
      "epoch": 0.823170731707317,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004938525371984588,
      "loss": 5.173,
      "mean_token_accuracy": 0.19178992211818696,
      "num_tokens": 18147173.0,
      "step": 9855
    },
    {
      "entropy": 5.267104911804199,
      "epoch": 0.8235883728700301,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004938456239263085,
      "loss": 5.162,
      "mean_token_accuracy": 0.19237794429063798,
      "num_tokens": 18157061.0,
      "step": 9860
    },
    {
      "entropy": 5.320634126663208,
      "epoch": 0.8240060140327431,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004938387068230021,
      "loss": 5.0933,
      "mean_token_accuracy": 0.19333020895719527,
      "num_tokens": 18166115.0,
      "step": 9865
    },
    {
      "entropy": 5.216364765167237,
      "epoch": 0.824423655195456,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004938317858886606,
      "loss": 5.1077,
      "mean_token_accuracy": 0.1979804202914238,
      "num_tokens": 18174978.0,
      "step": 9870
    },
    {
      "entropy": 5.291637516021728,
      "epoch": 0.824841296358169,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004938248611234055,
      "loss": 5.1905,
      "mean_token_accuracy": 0.19106807708740234,
      "num_tokens": 18184948.0,
      "step": 9875
    },
    {
      "entropy": 5.315226459503174,
      "epoch": 0.8252589375208821,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004938179325273577,
      "loss": 5.1204,
      "mean_token_accuracy": 0.19373727589845657,
      "num_tokens": 18193443.0,
      "step": 9880
    },
    {
      "entropy": 5.301239633560181,
      "epoch": 0.825676578683595,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004938110001006385,
      "loss": 5.1266,
      "mean_token_accuracy": 0.19410819411277772,
      "num_tokens": 18202512.0,
      "step": 9885
    },
    {
      "entropy": 5.300207042694092,
      "epoch": 0.826094219846308,
      "grad_norm": 1.125,
      "learning_rate": 0.0004938040638433694,
      "loss": 5.1582,
      "mean_token_accuracy": 0.19214901626110076,
      "num_tokens": 18212328.0,
      "step": 9890
    },
    {
      "entropy": 5.247486257553101,
      "epoch": 0.8265118610090211,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004937971237556719,
      "loss": 5.1066,
      "mean_token_accuracy": 0.19570307582616805,
      "num_tokens": 18221155.0,
      "step": 9895
    },
    {
      "entropy": 5.253676795959473,
      "epoch": 0.826929502171734,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004937901798376672,
      "loss": 5.1714,
      "mean_token_accuracy": 0.1919768840074539,
      "num_tokens": 18230604.0,
      "step": 9900
    },
    {
      "entropy": 5.271462249755859,
      "epoch": 0.827347143334447,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004937832320894772,
      "loss": 5.0957,
      "mean_token_accuracy": 0.1914784461259842,
      "num_tokens": 18239126.0,
      "step": 9905
    },
    {
      "entropy": 5.254875421524048,
      "epoch": 0.8277647844971601,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004937762805112233,
      "loss": 5.1346,
      "mean_token_accuracy": 0.18713102638721466,
      "num_tokens": 18248634.0,
      "step": 9910
    },
    {
      "entropy": 5.305291795730591,
      "epoch": 0.828182425659873,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004937693251030273,
      "loss": 5.2877,
      "mean_token_accuracy": 0.1792468711733818,
      "num_tokens": 18258500.0,
      "step": 9915
    },
    {
      "entropy": 5.340764713287354,
      "epoch": 0.828600066822586,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000493762365865011,
      "loss": 5.0887,
      "mean_token_accuracy": 0.19513607025146484,
      "num_tokens": 18268374.0,
      "step": 9920
    },
    {
      "entropy": 5.266375398635864,
      "epoch": 0.8290177079852991,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004937554027972961,
      "loss": 5.2041,
      "mean_token_accuracy": 0.18766897469758986,
      "num_tokens": 18277667.0,
      "step": 9925
    },
    {
      "entropy": 5.3305596828460695,
      "epoch": 0.829435349148012,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004937484359000046,
      "loss": 5.2268,
      "mean_token_accuracy": 0.185519739985466,
      "num_tokens": 18287516.0,
      "step": 9930
    },
    {
      "entropy": 5.2978802680969235,
      "epoch": 0.829852990310725,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004937414651732584,
      "loss": 5.0721,
      "mean_token_accuracy": 0.19688668251037597,
      "num_tokens": 18296534.0,
      "step": 9935
    },
    {
      "entropy": 5.289247226715088,
      "epoch": 0.830270631473438,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004937344906171795,
      "loss": 5.1988,
      "mean_token_accuracy": 0.19557883888483046,
      "num_tokens": 18305743.0,
      "step": 9940
    },
    {
      "entropy": 5.31955394744873,
      "epoch": 0.830688272636151,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004937275122318902,
      "loss": 5.2184,
      "mean_token_accuracy": 0.18593123108148574,
      "num_tokens": 18315524.0,
      "step": 9945
    },
    {
      "entropy": 5.250659704208374,
      "epoch": 0.831105913798864,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004937205300175123,
      "loss": 5.0729,
      "mean_token_accuracy": 0.19849158078432083,
      "num_tokens": 18324280.0,
      "step": 9950
    },
    {
      "entropy": 5.347445631027222,
      "epoch": 0.831523554961577,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004937135439741682,
      "loss": 5.172,
      "mean_token_accuracy": 0.1910255119204521,
      "num_tokens": 18332661.0,
      "step": 9955
    },
    {
      "entropy": 5.313532590866089,
      "epoch": 0.83194119612429,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004937065541019806,
      "loss": 5.2866,
      "mean_token_accuracy": 0.1713092193007469,
      "num_tokens": 18342986.0,
      "step": 9960
    },
    {
      "entropy": 5.348479986190796,
      "epoch": 0.832358837287003,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004936995604010711,
      "loss": 5.2232,
      "mean_token_accuracy": 0.18682226538658142,
      "num_tokens": 18351873.0,
      "step": 9965
    },
    {
      "entropy": 5.437933540344238,
      "epoch": 0.832776478449716,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004936925628715628,
      "loss": 5.2981,
      "mean_token_accuracy": 0.1859834834933281,
      "num_tokens": 18362174.0,
      "step": 9970
    },
    {
      "entropy": 5.267879819869995,
      "epoch": 0.833194119612429,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004936855615135779,
      "loss": 5.0765,
      "mean_token_accuracy": 0.1976146012544632,
      "num_tokens": 18371272.0,
      "step": 9975
    },
    {
      "entropy": 5.2829649448394775,
      "epoch": 0.833611760775142,
      "grad_norm": 1.4375,
      "learning_rate": 0.000493678556327239,
      "loss": 5.2101,
      "mean_token_accuracy": 0.1941166803240776,
      "num_tokens": 18380526.0,
      "step": 9980
    },
    {
      "entropy": 5.281093978881836,
      "epoch": 0.834029401937855,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004936715473126688,
      "loss": 5.1137,
      "mean_token_accuracy": 0.18696612268686294,
      "num_tokens": 18390027.0,
      "step": 9985
    },
    {
      "entropy": 5.236103534698486,
      "epoch": 0.8344470431005679,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00049366453446999,
      "loss": 5.1152,
      "mean_token_accuracy": 0.1956520676612854,
      "num_tokens": 18399012.0,
      "step": 9990
    },
    {
      "entropy": 5.1832630157470705,
      "epoch": 0.834864684263281,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004936575177993251,
      "loss": 5.0343,
      "mean_token_accuracy": 0.19478585869073867,
      "num_tokens": 18408891.0,
      "step": 9995
    },
    {
      "entropy": 5.227372360229492,
      "epoch": 0.835282325425994,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004936504973007973,
      "loss": 5.0713,
      "mean_token_accuracy": 0.19941119700670243,
      "num_tokens": 18417909.0,
      "step": 10000
    },
    {
      "entropy": 5.241479349136353,
      "epoch": 0.8356999665887069,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004936434729745294,
      "loss": 5.1396,
      "mean_token_accuracy": 0.1920836716890335,
      "num_tokens": 18427250.0,
      "step": 10005
    },
    {
      "entropy": 5.326820135116577,
      "epoch": 0.83611760775142,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004936364448206442,
      "loss": 5.1354,
      "mean_token_accuracy": 0.19181458950042723,
      "num_tokens": 18435851.0,
      "step": 10010
    },
    {
      "entropy": 5.214692258834839,
      "epoch": 0.836535248914133,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004936294128392649,
      "loss": 5.1236,
      "mean_token_accuracy": 0.18665034770965577,
      "num_tokens": 18444535.0,
      "step": 10015
    },
    {
      "entropy": 5.299440240859985,
      "epoch": 0.8369528900768459,
      "grad_norm": 1.125,
      "learning_rate": 0.0004936223770305146,
      "loss": 5.2469,
      "mean_token_accuracy": 0.18339482694864273,
      "num_tokens": 18454453.0,
      "step": 10020
    },
    {
      "entropy": 5.449494171142578,
      "epoch": 0.837370531239559,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004936153373945165,
      "loss": 5.2788,
      "mean_token_accuracy": 0.18121690005064012,
      "num_tokens": 18465033.0,
      "step": 10025
    },
    {
      "entropy": 5.285546445846558,
      "epoch": 0.837788172402272,
      "grad_norm": 1.25,
      "learning_rate": 0.0004936082939313936,
      "loss": 5.1609,
      "mean_token_accuracy": 0.19212604314088821,
      "num_tokens": 18473786.0,
      "step": 10030
    },
    {
      "entropy": 5.19959020614624,
      "epoch": 0.8382058135649849,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004936012466412695,
      "loss": 5.1171,
      "mean_token_accuracy": 0.1842856928706169,
      "num_tokens": 18483282.0,
      "step": 10035
    },
    {
      "entropy": 5.40060658454895,
      "epoch": 0.838623454727698,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004935941955242674,
      "loss": 5.2813,
      "mean_token_accuracy": 0.1879670113325119,
      "num_tokens": 18492612.0,
      "step": 10040
    },
    {
      "entropy": 5.28286919593811,
      "epoch": 0.839041095890411,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004935871405805109,
      "loss": 5.0885,
      "mean_token_accuracy": 0.2009210392832756,
      "num_tokens": 18501120.0,
      "step": 10045
    },
    {
      "entropy": 5.261589527130127,
      "epoch": 0.8394587370531239,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004935800818101232,
      "loss": 5.2166,
      "mean_token_accuracy": 0.18312245309352876,
      "num_tokens": 18510613.0,
      "step": 10050
    },
    {
      "entropy": 5.291419506072998,
      "epoch": 0.8398763782158369,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004935730192132282,
      "loss": 5.0922,
      "mean_token_accuracy": 0.19438743889331817,
      "num_tokens": 18518854.0,
      "step": 10055
    },
    {
      "entropy": 5.300976705551148,
      "epoch": 0.84029401937855,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004935659527899493,
      "loss": 5.1666,
      "mean_token_accuracy": 0.193764728307724,
      "num_tokens": 18527592.0,
      "step": 10060
    },
    {
      "entropy": 5.229451322555542,
      "epoch": 0.8407116605412629,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004935588825404104,
      "loss": 5.0543,
      "mean_token_accuracy": 0.19931281059980394,
      "num_tokens": 18536396.0,
      "step": 10065
    },
    {
      "entropy": 5.232517862319947,
      "epoch": 0.8411293017039759,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004935518084647351,
      "loss": 5.0287,
      "mean_token_accuracy": 0.20150027871131898,
      "num_tokens": 18545201.0,
      "step": 10070
    },
    {
      "entropy": 5.242734289169311,
      "epoch": 0.841546942866689,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004935447305630474,
      "loss": 5.2227,
      "mean_token_accuracy": 0.18361724764108658,
      "num_tokens": 18554944.0,
      "step": 10075
    },
    {
      "entropy": 5.234609651565552,
      "epoch": 0.8419645840294019,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004935376488354711,
      "loss": 5.1015,
      "mean_token_accuracy": 0.19018730223178865,
      "num_tokens": 18564642.0,
      "step": 10080
    },
    {
      "entropy": 5.271942758560181,
      "epoch": 0.8423822251921149,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004935305632821303,
      "loss": 5.1176,
      "mean_token_accuracy": 0.19007254838943483,
      "num_tokens": 18573897.0,
      "step": 10085
    },
    {
      "entropy": 5.261401033401489,
      "epoch": 0.842799866354828,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004935234739031488,
      "loss": 5.1555,
      "mean_token_accuracy": 0.19369743764400482,
      "num_tokens": 18583174.0,
      "step": 10090
    },
    {
      "entropy": 5.278877210617066,
      "epoch": 0.8432175075175409,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004935163806986508,
      "loss": 5.0623,
      "mean_token_accuracy": 0.19669793546199799,
      "num_tokens": 18592540.0,
      "step": 10095
    },
    {
      "entropy": 5.205090188980103,
      "epoch": 0.8436351486802539,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004935092836687607,
      "loss": 5.075,
      "mean_token_accuracy": 0.19474110305309295,
      "num_tokens": 18602743.0,
      "step": 10100
    },
    {
      "entropy": 5.2059714794158936,
      "epoch": 0.8440527898429669,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004935021828136025,
      "loss": 5.0545,
      "mean_token_accuracy": 0.19375797510147094,
      "num_tokens": 18611544.0,
      "step": 10105
    },
    {
      "entropy": 5.307067632675171,
      "epoch": 0.8444704310056799,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004934950781333005,
      "loss": 5.1488,
      "mean_token_accuracy": 0.19102364480495454,
      "num_tokens": 18620841.0,
      "step": 10110
    },
    {
      "entropy": 5.2926441669464115,
      "epoch": 0.8448880721683929,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004934879696279793,
      "loss": 5.2014,
      "mean_token_accuracy": 0.18762198835611343,
      "num_tokens": 18630890.0,
      "step": 10115
    },
    {
      "entropy": 5.3619307518005375,
      "epoch": 0.8453057133311059,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004934808572977632,
      "loss": 5.1998,
      "mean_token_accuracy": 0.18724320232868194,
      "num_tokens": 18639768.0,
      "step": 10120
    },
    {
      "entropy": 5.191465616226196,
      "epoch": 0.8457233544938189,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004934737411427765,
      "loss": 5.0173,
      "mean_token_accuracy": 0.19451924413442612,
      "num_tokens": 18649578.0,
      "step": 10125
    },
    {
      "entropy": 5.192807817459107,
      "epoch": 0.8461409956565319,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004934666211631442,
      "loss": 5.1166,
      "mean_token_accuracy": 0.19320215433835983,
      "num_tokens": 18658470.0,
      "step": 10130
    },
    {
      "entropy": 5.2549967765808105,
      "epoch": 0.8465586368192449,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004934594973589906,
      "loss": 5.0857,
      "mean_token_accuracy": 0.18834662139415742,
      "num_tokens": 18668037.0,
      "step": 10135
    },
    {
      "entropy": 5.296866321563721,
      "epoch": 0.8469762779819578,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004934523697304406,
      "loss": 5.1999,
      "mean_token_accuracy": 0.1879106268286705,
      "num_tokens": 18677841.0,
      "step": 10140
    },
    {
      "entropy": 5.327513980865478,
      "epoch": 0.8473939191446709,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004934452382776189,
      "loss": 5.1552,
      "mean_token_accuracy": 0.1882873684167862,
      "num_tokens": 18686244.0,
      "step": 10145
    },
    {
      "entropy": 5.21918830871582,
      "epoch": 0.8478115603073839,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004934381030006504,
      "loss": 5.1894,
      "mean_token_accuracy": 0.19015713185071945,
      "num_tokens": 18695129.0,
      "step": 10150
    },
    {
      "entropy": 5.1736023902893065,
      "epoch": 0.8482292014700968,
      "grad_norm": 1.453125,
      "learning_rate": 0.00049343096389966,
      "loss": 4.995,
      "mean_token_accuracy": 0.2052477166056633,
      "num_tokens": 18703566.0,
      "step": 10155
    },
    {
      "entropy": 5.357991790771484,
      "epoch": 0.8486468426328099,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004934238209747724,
      "loss": 5.215,
      "mean_token_accuracy": 0.18315352350473404,
      "num_tokens": 18713616.0,
      "step": 10160
    },
    {
      "entropy": 5.359310340881348,
      "epoch": 0.8490644837955229,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004934166742261132,
      "loss": 5.1389,
      "mean_token_accuracy": 0.18987988978624343,
      "num_tokens": 18722851.0,
      "step": 10165
    },
    {
      "entropy": 5.269087791442871,
      "epoch": 0.8494821249582358,
      "grad_norm": 1.34375,
      "learning_rate": 0.000493409523653807,
      "loss": 5.1424,
      "mean_token_accuracy": 0.19145039319992066,
      "num_tokens": 18731961.0,
      "step": 10170
    },
    {
      "entropy": 5.313898181915283,
      "epoch": 0.8498997661209489,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004934023692579793,
      "loss": 5.3135,
      "mean_token_accuracy": 0.18967837542295457,
      "num_tokens": 18741267.0,
      "step": 10175
    },
    {
      "entropy": 5.266151666641235,
      "epoch": 0.8503174072836619,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004933952110387553,
      "loss": 5.1429,
      "mean_token_accuracy": 0.19029432982206346,
      "num_tokens": 18750266.0,
      "step": 10180
    },
    {
      "entropy": 5.254126405715942,
      "epoch": 0.8507350484463748,
      "grad_norm": 1.25,
      "learning_rate": 0.0004933880489962602,
      "loss": 5.1384,
      "mean_token_accuracy": 0.1929827407002449,
      "num_tokens": 18759910.0,
      "step": 10185
    },
    {
      "entropy": 5.419958305358887,
      "epoch": 0.8511526896090879,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004933808831306194,
      "loss": 5.2771,
      "mean_token_accuracy": 0.18061521649360657,
      "num_tokens": 18769510.0,
      "step": 10190
    },
    {
      "entropy": 5.254436063766479,
      "epoch": 0.8515703307718009,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004933737134419583,
      "loss": 5.0819,
      "mean_token_accuracy": 0.19432806670665742,
      "num_tokens": 18779893.0,
      "step": 10195
    },
    {
      "entropy": 5.295120620727539,
      "epoch": 0.8519879719345138,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004933665399304025,
      "loss": 5.1381,
      "mean_token_accuracy": 0.19522347599267958,
      "num_tokens": 18788802.0,
      "step": 10200
    },
    {
      "entropy": 5.200299406051636,
      "epoch": 0.8524056130972268,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004933593625960777,
      "loss": 5.0503,
      "mean_token_accuracy": 0.19243860393762588,
      "num_tokens": 18797154.0,
      "step": 10205
    },
    {
      "entropy": 5.2647655487060545,
      "epoch": 0.8528232542599399,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004933521814391094,
      "loss": 5.164,
      "mean_token_accuracy": 0.18749276101589202,
      "num_tokens": 18806151.0,
      "step": 10210
    },
    {
      "entropy": 5.381960535049439,
      "epoch": 0.8532408954226529,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004933449964596232,
      "loss": 5.0737,
      "mean_token_accuracy": 0.19823941290378572,
      "num_tokens": 18815162.0,
      "step": 10215
    },
    {
      "entropy": 5.174137926101684,
      "epoch": 0.8536585365853658,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004933378076577451,
      "loss": 5.0738,
      "mean_token_accuracy": 0.1869513675570488,
      "num_tokens": 18823676.0,
      "step": 10220
    },
    {
      "entropy": 5.311799240112305,
      "epoch": 0.8540761777480789,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004933306150336009,
      "loss": 5.3526,
      "mean_token_accuracy": 0.18070393651723862,
      "num_tokens": 18832991.0,
      "step": 10225
    },
    {
      "entropy": 5.4486793041229244,
      "epoch": 0.8544938189107919,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004933234185873164,
      "loss": 5.2199,
      "mean_token_accuracy": 0.1883494734764099,
      "num_tokens": 18842614.0,
      "step": 10230
    },
    {
      "entropy": 5.437237167358399,
      "epoch": 0.8549114600735048,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004933162183190177,
      "loss": 5.2853,
      "mean_token_accuracy": 0.1798120841383934,
      "num_tokens": 18852747.0,
      "step": 10235
    },
    {
      "entropy": 5.226648139953613,
      "epoch": 0.8553291012362179,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004933090142288308,
      "loss": 5.1587,
      "mean_token_accuracy": 0.19008901119232177,
      "num_tokens": 18862005.0,
      "step": 10240
    },
    {
      "entropy": 5.255118417739868,
      "epoch": 0.8557467423989309,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004933018063168819,
      "loss": 5.0954,
      "mean_token_accuracy": 0.20613744854927063,
      "num_tokens": 18870172.0,
      "step": 10245
    },
    {
      "entropy": 5.286107587814331,
      "epoch": 0.8561643835616438,
      "grad_norm": 1.234375,
      "learning_rate": 0.000493294594583297,
      "loss": 5.1629,
      "mean_token_accuracy": 0.1901986002922058,
      "num_tokens": 18879428.0,
      "step": 10250
    },
    {
      "entropy": 5.255838537216187,
      "epoch": 0.8565820247243568,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004932873790282026,
      "loss": 5.1135,
      "mean_token_accuracy": 0.20238856673240663,
      "num_tokens": 18888694.0,
      "step": 10255
    },
    {
      "entropy": 5.367196941375733,
      "epoch": 0.8569996658870699,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004932801596517248,
      "loss": 5.183,
      "mean_token_accuracy": 0.18612778186798096,
      "num_tokens": 18898317.0,
      "step": 10260
    },
    {
      "entropy": 5.260575437545777,
      "epoch": 0.8574173070497828,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004932729364539901,
      "loss": 5.1498,
      "mean_token_accuracy": 0.18709730356931686,
      "num_tokens": 18906749.0,
      "step": 10265
    },
    {
      "entropy": 5.220563507080078,
      "epoch": 0.8578349482124958,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004932657094351248,
      "loss": 5.0564,
      "mean_token_accuracy": 0.19547294229269027,
      "num_tokens": 18915404.0,
      "step": 10270
    },
    {
      "entropy": 5.254152107238769,
      "epoch": 0.8582525893752089,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004932584785952555,
      "loss": 5.1587,
      "mean_token_accuracy": 0.19268184453248977,
      "num_tokens": 18925512.0,
      "step": 10275
    },
    {
      "entropy": 5.2647998332977295,
      "epoch": 0.8586702305379218,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000493251243934509,
      "loss": 5.0892,
      "mean_token_accuracy": 0.19869595319032668,
      "num_tokens": 18934278.0,
      "step": 10280
    },
    {
      "entropy": 5.3008111953735355,
      "epoch": 0.8590878717006348,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004932440054530116,
      "loss": 5.1569,
      "mean_token_accuracy": 0.19053591787815094,
      "num_tokens": 18943126.0,
      "step": 10285
    },
    {
      "entropy": 5.301484870910644,
      "epoch": 0.8595055128633479,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004932367631508903,
      "loss": 5.1555,
      "mean_token_accuracy": 0.19479513466358184,
      "num_tokens": 18951716.0,
      "step": 10290
    },
    {
      "entropy": 5.30456428527832,
      "epoch": 0.8599231540260608,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004932295170282717,
      "loss": 5.2375,
      "mean_token_accuracy": 0.18485366106033324,
      "num_tokens": 18960692.0,
      "step": 10295
    },
    {
      "entropy": 5.289656782150269,
      "epoch": 0.8603407951887738,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004932222670852828,
      "loss": 5.1671,
      "mean_token_accuracy": 0.19203932881355285,
      "num_tokens": 18969813.0,
      "step": 10300
    },
    {
      "entropy": 5.286883211135864,
      "epoch": 0.8607584363514869,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004932150133220504,
      "loss": 5.037,
      "mean_token_accuracy": 0.19790283888578414,
      "num_tokens": 18978536.0,
      "step": 10305
    },
    {
      "entropy": 5.239656019210815,
      "epoch": 0.8611760775141998,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004932077557387015,
      "loss": 5.1164,
      "mean_token_accuracy": 0.19691624790430068,
      "num_tokens": 18987637.0,
      "step": 10310
    },
    {
      "entropy": 5.148351049423217,
      "epoch": 0.8615937186769128,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004932004943353633,
      "loss": 4.9524,
      "mean_token_accuracy": 0.2002424567937851,
      "num_tokens": 18997010.0,
      "step": 10315
    },
    {
      "entropy": 5.264727306365967,
      "epoch": 0.8620113598396258,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004931932291121627,
      "loss": 5.1208,
      "mean_token_accuracy": 0.1872481882572174,
      "num_tokens": 19005685.0,
      "step": 10320
    },
    {
      "entropy": 5.237448215484619,
      "epoch": 0.8624290010023388,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004931859600692269,
      "loss": 5.1166,
      "mean_token_accuracy": 0.1944681003689766,
      "num_tokens": 19014836.0,
      "step": 10325
    },
    {
      "entropy": 5.2952064037322994,
      "epoch": 0.8628466421650518,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004931786872066833,
      "loss": 5.1567,
      "mean_token_accuracy": 0.1878812938928604,
      "num_tokens": 19024442.0,
      "step": 10330
    },
    {
      "entropy": 5.242105388641358,
      "epoch": 0.8632642833277648,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004931714105246592,
      "loss": 5.1104,
      "mean_token_accuracy": 0.19340454638004304,
      "num_tokens": 19033730.0,
      "step": 10335
    },
    {
      "entropy": 5.250036001205444,
      "epoch": 0.8636819244904778,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004931641300232819,
      "loss": 5.0675,
      "mean_token_accuracy": 0.20513610541820526,
      "num_tokens": 19042765.0,
      "step": 10340
    },
    {
      "entropy": 5.281334924697876,
      "epoch": 0.8640995656531908,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004931568457026789,
      "loss": 5.0767,
      "mean_token_accuracy": 0.18846557140350342,
      "num_tokens": 19051408.0,
      "step": 10345
    },
    {
      "entropy": 5.296620655059814,
      "epoch": 0.8645172068159038,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004931495575629778,
      "loss": 5.189,
      "mean_token_accuracy": 0.18550419956445693,
      "num_tokens": 19061145.0,
      "step": 10350
    },
    {
      "entropy": 5.272748708724976,
      "epoch": 0.8649348479786167,
      "grad_norm": 1.234375,
      "learning_rate": 0.000493142265604306,
      "loss": 5.1858,
      "mean_token_accuracy": 0.18847290873527528,
      "num_tokens": 19070157.0,
      "step": 10355
    },
    {
      "entropy": 5.2359295845031735,
      "epoch": 0.8653524891413298,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004931349698267914,
      "loss": 4.9977,
      "mean_token_accuracy": 0.20588619858026505,
      "num_tokens": 19078532.0,
      "step": 10360
    },
    {
      "entropy": 5.22608208656311,
      "epoch": 0.8657701303040428,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004931276702305615,
      "loss": 5.0712,
      "mean_token_accuracy": 0.19595552384853362,
      "num_tokens": 19087283.0,
      "step": 10365
    },
    {
      "entropy": 5.25261754989624,
      "epoch": 0.8661877714667557,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004931203668157441,
      "loss": 5.2251,
      "mean_token_accuracy": 0.1898394986987114,
      "num_tokens": 19097568.0,
      "step": 10370
    },
    {
      "entropy": 5.3318439483642575,
      "epoch": 0.8666054126294688,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004931130595824672,
      "loss": 5.1908,
      "mean_token_accuracy": 0.183209528028965,
      "num_tokens": 19107218.0,
      "step": 10375
    },
    {
      "entropy": 5.302245140075684,
      "epoch": 0.8670230537921818,
      "grad_norm": 1.375,
      "learning_rate": 0.0004931057485308587,
      "loss": 5.1572,
      "mean_token_accuracy": 0.19403398633003235,
      "num_tokens": 19116128.0,
      "step": 10380
    },
    {
      "entropy": 5.211310148239136,
      "epoch": 0.8674406949548947,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004930984336610465,
      "loss": 5.1272,
      "mean_token_accuracy": 0.1948370084166527,
      "num_tokens": 19124885.0,
      "step": 10385
    },
    {
      "entropy": 5.340264129638672,
      "epoch": 0.8678583361176078,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004930911149731587,
      "loss": 5.1949,
      "mean_token_accuracy": 0.19367161989212037,
      "num_tokens": 19134125.0,
      "step": 10390
    },
    {
      "entropy": 5.259817314147949,
      "epoch": 0.8682759772803208,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004930837924673233,
      "loss": 5.0752,
      "mean_token_accuracy": 0.1934529200196266,
      "num_tokens": 19141925.0,
      "step": 10395
    },
    {
      "entropy": 5.279439973831177,
      "epoch": 0.8686936184430337,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004930764661436688,
      "loss": 5.1623,
      "mean_token_accuracy": 0.19201445579528809,
      "num_tokens": 19151440.0,
      "step": 10400
    },
    {
      "entropy": 5.300766372680664,
      "epoch": 0.8691112596057468,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004930691360023231,
      "loss": 5.2421,
      "mean_token_accuracy": 0.1950339764356613,
      "num_tokens": 19161447.0,
      "step": 10405
    },
    {
      "entropy": 5.340952634811401,
      "epoch": 0.8695289007684598,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004930618020434147,
      "loss": 5.1634,
      "mean_token_accuracy": 0.19083169102668762,
      "num_tokens": 19170804.0,
      "step": 10410
    },
    {
      "entropy": 5.408653211593628,
      "epoch": 0.8699465419311727,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004930544642670721,
      "loss": 5.2853,
      "mean_token_accuracy": 0.1818897932767868,
      "num_tokens": 19179898.0,
      "step": 10415
    },
    {
      "entropy": 5.281631326675415,
      "epoch": 0.8703641830938857,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004930471226734235,
      "loss": 5.0882,
      "mean_token_accuracy": 0.1962774321436882,
      "num_tokens": 19188919.0,
      "step": 10420
    },
    {
      "entropy": 5.275079774856567,
      "epoch": 0.8707818242565988,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004930397772625976,
      "loss": 5.1791,
      "mean_token_accuracy": 0.18957459479570388,
      "num_tokens": 19197745.0,
      "step": 10425
    },
    {
      "entropy": 5.20112156867981,
      "epoch": 0.8711994654193117,
      "grad_norm": 1.296875,
      "learning_rate": 0.000493032428034723,
      "loss": 5.179,
      "mean_token_accuracy": 0.19445342272520066,
      "num_tokens": 19206681.0,
      "step": 10430
    },
    {
      "entropy": 5.361344575881958,
      "epoch": 0.8716171065820247,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004930250749899282,
      "loss": 5.1671,
      "mean_token_accuracy": 0.1909335732460022,
      "num_tokens": 19215806.0,
      "step": 10435
    },
    {
      "entropy": 5.33047046661377,
      "epoch": 0.8720347477447378,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004930177181283422,
      "loss": 5.1894,
      "mean_token_accuracy": 0.1892147734761238,
      "num_tokens": 19224681.0,
      "step": 10440
    },
    {
      "entropy": 5.274596166610718,
      "epoch": 0.8724523889074507,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004930103574500934,
      "loss": 5.1769,
      "mean_token_accuracy": 0.19144096672534944,
      "num_tokens": 19232684.0,
      "step": 10445
    },
    {
      "entropy": 5.313355541229248,
      "epoch": 0.8728700300701637,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004930029929553109,
      "loss": 5.0996,
      "mean_token_accuracy": 0.18856192082166673,
      "num_tokens": 19242538.0,
      "step": 10450
    },
    {
      "entropy": 5.192029285430908,
      "epoch": 0.8732876712328768,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004929956246441237,
      "loss": 5.1337,
      "mean_token_accuracy": 0.18757807910442353,
      "num_tokens": 19251146.0,
      "step": 10455
    },
    {
      "entropy": 5.362995862960815,
      "epoch": 0.8737053123955897,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004929882525166608,
      "loss": 5.1838,
      "mean_token_accuracy": 0.1886983186006546,
      "num_tokens": 19259931.0,
      "step": 10460
    },
    {
      "entropy": 5.32528395652771,
      "epoch": 0.8741229535583027,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004929808765730509,
      "loss": 5.178,
      "mean_token_accuracy": 0.18712896555662156,
      "num_tokens": 19268796.0,
      "step": 10465
    },
    {
      "entropy": 5.228519105911255,
      "epoch": 0.8745405947210158,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004929734968134235,
      "loss": 5.1911,
      "mean_token_accuracy": 0.19131958931684495,
      "num_tokens": 19278396.0,
      "step": 10470
    },
    {
      "entropy": 5.318662118911743,
      "epoch": 0.8749582358837287,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004929661132379076,
      "loss": 5.217,
      "mean_token_accuracy": 0.18495183885097505,
      "num_tokens": 19287914.0,
      "step": 10475
    },
    {
      "entropy": 5.325718545913697,
      "epoch": 0.8753758770464417,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004929587258466326,
      "loss": 5.1856,
      "mean_token_accuracy": 0.1836814597249031,
      "num_tokens": 19297304.0,
      "step": 10480
    },
    {
      "entropy": 5.311287260055542,
      "epoch": 0.8757935182091547,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004929513346397278,
      "loss": 5.1277,
      "mean_token_accuracy": 0.19700873494148255,
      "num_tokens": 19306245.0,
      "step": 10485
    },
    {
      "entropy": 5.262881565093994,
      "epoch": 0.8762111593718677,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004929439396173223,
      "loss": 5.1496,
      "mean_token_accuracy": 0.1845311090350151,
      "num_tokens": 19316160.0,
      "step": 10490
    },
    {
      "entropy": 5.250891065597534,
      "epoch": 0.8766288005345807,
      "grad_norm": 1.3125,
      "learning_rate": 0.000492936540779546,
      "loss": 5.2103,
      "mean_token_accuracy": 0.18243358582258223,
      "num_tokens": 19325411.0,
      "step": 10495
    },
    {
      "entropy": 5.369228219985962,
      "epoch": 0.8770464416972937,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004929291381265281,
      "loss": 5.1375,
      "mean_token_accuracy": 0.1911255046725273,
      "num_tokens": 19334297.0,
      "step": 10500
    },
    {
      "entropy": 5.3839537620544435,
      "epoch": 0.8774640828600067,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004929217316583985,
      "loss": 5.2403,
      "mean_token_accuracy": 0.1834528774023056,
      "num_tokens": 19344121.0,
      "step": 10505
    },
    {
      "entropy": 5.280856418609619,
      "epoch": 0.8778817240227197,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004929143213752865,
      "loss": 5.1633,
      "mean_token_accuracy": 0.18882547318935394,
      "num_tokens": 19353428.0,
      "step": 10510
    },
    {
      "entropy": 5.3072315692901615,
      "epoch": 0.8782993651854327,
      "grad_norm": 1.453125,
      "learning_rate": 0.000492906907277322,
      "loss": 5.221,
      "mean_token_accuracy": 0.18923131227493287,
      "num_tokens": 19362661.0,
      "step": 10515
    },
    {
      "entropy": 5.1896954536437985,
      "epoch": 0.8787170063481456,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000492899489364635,
      "loss": 5.0218,
      "mean_token_accuracy": 0.2030668392777443,
      "num_tokens": 19371246.0,
      "step": 10520
    },
    {
      "entropy": 5.2653402328491214,
      "epoch": 0.8791346475108587,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004928920676373549,
      "loss": 5.0833,
      "mean_token_accuracy": 0.1945516958832741,
      "num_tokens": 19379777.0,
      "step": 10525
    },
    {
      "entropy": 5.23420672416687,
      "epoch": 0.8795522886735717,
      "grad_norm": 1.21875,
      "learning_rate": 0.000492884642095612,
      "loss": 5.1099,
      "mean_token_accuracy": 0.18972722589969634,
      "num_tokens": 19388717.0,
      "step": 10530
    },
    {
      "entropy": 5.433230257034301,
      "epoch": 0.8799699298362846,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004928772127395361,
      "loss": 5.2382,
      "mean_token_accuracy": 0.18339210748672485,
      "num_tokens": 19397936.0,
      "step": 10535
    },
    {
      "entropy": 5.388091611862182,
      "epoch": 0.8803875709989977,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004928697795692575,
      "loss": 5.21,
      "mean_token_accuracy": 0.17988236397504806,
      "num_tokens": 19406830.0,
      "step": 10540
    },
    {
      "entropy": 5.283012390136719,
      "epoch": 0.8808052121617107,
      "grad_norm": 1.484375,
      "learning_rate": 0.000492862342584906,
      "loss": 5.2253,
      "mean_token_accuracy": 0.18579084128141404,
      "num_tokens": 19415131.0,
      "step": 10545
    },
    {
      "entropy": 5.256882810592652,
      "epoch": 0.8812228533244236,
      "grad_norm": 1.28125,
      "learning_rate": 0.000492854901786612,
      "loss": 5.0068,
      "mean_token_accuracy": 0.20541125684976577,
      "num_tokens": 19423585.0,
      "step": 10550
    },
    {
      "entropy": 5.191609525680542,
      "epoch": 0.8816404944871367,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004928474571745058,
      "loss": 5.0308,
      "mean_token_accuracy": 0.20325697362422943,
      "num_tokens": 19431890.0,
      "step": 10555
    },
    {
      "entropy": 5.297631454467774,
      "epoch": 0.8820581356498497,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004928400087487175,
      "loss": 5.1787,
      "mean_token_accuracy": 0.1893961176276207,
      "num_tokens": 19441377.0,
      "step": 10560
    },
    {
      "entropy": 5.3560121059417725,
      "epoch": 0.8824757768125626,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004928325565093777,
      "loss": 5.1711,
      "mean_token_accuracy": 0.18682490885257722,
      "num_tokens": 19451003.0,
      "step": 10565
    },
    {
      "entropy": 5.201437282562256,
      "epoch": 0.8828934179752757,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004928251004566167,
      "loss": 5.0682,
      "mean_token_accuracy": 0.19823963195085526,
      "num_tokens": 19460166.0,
      "step": 10570
    },
    {
      "entropy": 5.225899171829224,
      "epoch": 0.8833110591379887,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004928176405905653,
      "loss": 5.1616,
      "mean_token_accuracy": 0.194552044570446,
      "num_tokens": 19469417.0,
      "step": 10575
    },
    {
      "entropy": 5.332932424545288,
      "epoch": 0.8837287003007016,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004928101769113537,
      "loss": 5.2049,
      "mean_token_accuracy": 0.19193755388259887,
      "num_tokens": 19479275.0,
      "step": 10580
    },
    {
      "entropy": 5.372676944732666,
      "epoch": 0.8841463414634146,
      "grad_norm": 1.4609375,
      "learning_rate": 0.000492802709419113,
      "loss": 5.1358,
      "mean_token_accuracy": 0.18758666068315505,
      "num_tokens": 19488804.0,
      "step": 10585
    },
    {
      "entropy": 5.337365436553955,
      "epoch": 0.8845639826261277,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004927952381139735,
      "loss": 5.0867,
      "mean_token_accuracy": 0.20051475763320922,
      "num_tokens": 19497701.0,
      "step": 10590
    },
    {
      "entropy": 5.297200012207031,
      "epoch": 0.8849816237888406,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004927877629960663,
      "loss": 5.1246,
      "mean_token_accuracy": 0.19307359606027602,
      "num_tokens": 19508177.0,
      "step": 10595
    },
    {
      "entropy": 5.233062219619751,
      "epoch": 0.8853992649515536,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004927802840655221,
      "loss": 5.1825,
      "mean_token_accuracy": 0.19335023313760757,
      "num_tokens": 19516866.0,
      "step": 10600
    },
    {
      "entropy": 5.320388984680176,
      "epoch": 0.8858169061142667,
      "grad_norm": 1.4453125,
      "learning_rate": 0.000492772801322472,
      "loss": 5.191,
      "mean_token_accuracy": 0.1868289053440094,
      "num_tokens": 19525788.0,
      "step": 10605
    },
    {
      "entropy": 5.272816467285156,
      "epoch": 0.8862345472769796,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004927653147670468,
      "loss": 5.0137,
      "mean_token_accuracy": 0.19312302619218827,
      "num_tokens": 19534748.0,
      "step": 10610
    },
    {
      "entropy": 5.211900949478149,
      "epoch": 0.8866521884396926,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004927578243993777,
      "loss": 5.0934,
      "mean_token_accuracy": 0.1995951399207115,
      "num_tokens": 19544083.0,
      "step": 10615
    },
    {
      "entropy": 5.213784646987915,
      "epoch": 0.8870698296024057,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004927503302195957,
      "loss": 5.1098,
      "mean_token_accuracy": 0.19440722465515137,
      "num_tokens": 19553978.0,
      "step": 10620
    },
    {
      "entropy": 5.319274663925171,
      "epoch": 0.8874874707651186,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000492742832227832,
      "loss": 5.1827,
      "mean_token_accuracy": 0.1927887201309204,
      "num_tokens": 19563265.0,
      "step": 10625
    },
    {
      "entropy": 5.368842554092407,
      "epoch": 0.8879051119278316,
      "grad_norm": 1.421875,
      "learning_rate": 0.000492735330424218,
      "loss": 5.2027,
      "mean_token_accuracy": 0.1767944350838661,
      "num_tokens": 19572904.0,
      "step": 10630
    },
    {
      "entropy": 5.340467834472657,
      "epoch": 0.8883227530905446,
      "grad_norm": 1.25,
      "learning_rate": 0.000492727824808885,
      "loss": 5.2058,
      "mean_token_accuracy": 0.1939498469233513,
      "num_tokens": 19581680.0,
      "step": 10635
    },
    {
      "entropy": 5.235644292831421,
      "epoch": 0.8887403942532576,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004927203153819643,
      "loss": 5.0698,
      "mean_token_accuracy": 0.19437360763549805,
      "num_tokens": 19590885.0,
      "step": 10640
    },
    {
      "entropy": 5.157550096511841,
      "epoch": 0.8891580354159706,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004927128021435875,
      "loss": 5.0711,
      "mean_token_accuracy": 0.19800787419080734,
      "num_tokens": 19599901.0,
      "step": 10645
    },
    {
      "entropy": 5.330171537399292,
      "epoch": 0.8895756765786836,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000492705285093886,
      "loss": 5.2014,
      "mean_token_accuracy": 0.18586840182542802,
      "num_tokens": 19608745.0,
      "step": 10650
    },
    {
      "entropy": 5.231226110458374,
      "epoch": 0.8899933177413966,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004926977642329913,
      "loss": 4.9878,
      "mean_token_accuracy": 0.19670885056257248,
      "num_tokens": 19618418.0,
      "step": 10655
    },
    {
      "entropy": 5.217711591720581,
      "epoch": 0.8904109589041096,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004926902395610353,
      "loss": 5.0766,
      "mean_token_accuracy": 0.1872383177280426,
      "num_tokens": 19627118.0,
      "step": 10660
    },
    {
      "entropy": 5.311285734176636,
      "epoch": 0.8908286000668226,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004926827110781496,
      "loss": 5.2154,
      "mean_token_accuracy": 0.18928681463003158,
      "num_tokens": 19636565.0,
      "step": 10665
    },
    {
      "entropy": 5.252695465087891,
      "epoch": 0.8912462412295356,
      "grad_norm": 1.5,
      "learning_rate": 0.000492675178784466,
      "loss": 5.0782,
      "mean_token_accuracy": 0.19249269664287566,
      "num_tokens": 19645038.0,
      "step": 10670
    },
    {
      "entropy": 5.242134380340576,
      "epoch": 0.8916638823922486,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004926676426801164,
      "loss": 5.1075,
      "mean_token_accuracy": 0.1895960360765457,
      "num_tokens": 19654088.0,
      "step": 10675
    },
    {
      "entropy": 5.286896848678589,
      "epoch": 0.8920815235549616,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004926601027652328,
      "loss": 5.1895,
      "mean_token_accuracy": 0.1915432706475258,
      "num_tokens": 19662551.0,
      "step": 10680
    },
    {
      "entropy": 5.36880202293396,
      "epoch": 0.8924991647176745,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000492652559039947,
      "loss": 5.1377,
      "mean_token_accuracy": 0.19205421060323716,
      "num_tokens": 19672405.0,
      "step": 10685
    },
    {
      "entropy": 5.204009485244751,
      "epoch": 0.8929168058803876,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004926450115043912,
      "loss": 5.1355,
      "mean_token_accuracy": 0.1928644061088562,
      "num_tokens": 19681288.0,
      "step": 10690
    },
    {
      "entropy": 5.223655939102173,
      "epoch": 0.8933344470431006,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004926374601586975,
      "loss": 5.058,
      "mean_token_accuracy": 0.19555283784866334,
      "num_tokens": 19690723.0,
      "step": 10695
    },
    {
      "entropy": 5.273682928085327,
      "epoch": 0.8937520882058135,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004926299050029981,
      "loss": 5.1053,
      "mean_token_accuracy": 0.19306541085243226,
      "num_tokens": 19699642.0,
      "step": 10700
    },
    {
      "entropy": 5.251811170578003,
      "epoch": 0.8941697293685266,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004926223460374251,
      "loss": 5.0562,
      "mean_token_accuracy": 0.19870668053627014,
      "num_tokens": 19707563.0,
      "step": 10705
    },
    {
      "entropy": 5.310912084579468,
      "epoch": 0.8945873705312396,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004926147832621111,
      "loss": 5.2113,
      "mean_token_accuracy": 0.1899917483329773,
      "num_tokens": 19717109.0,
      "step": 10710
    },
    {
      "entropy": 5.2104407787323,
      "epoch": 0.8950050116939525,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004926072166771884,
      "loss": 5.0325,
      "mean_token_accuracy": 0.20259149372577667,
      "num_tokens": 19725585.0,
      "step": 10715
    },
    {
      "entropy": 5.121189832687378,
      "epoch": 0.8954226528566656,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004925996462827895,
      "loss": 4.9872,
      "mean_token_accuracy": 0.2061747819185257,
      "num_tokens": 19733502.0,
      "step": 10720
    },
    {
      "entropy": 5.24937949180603,
      "epoch": 0.8958402940193786,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004925920720790466,
      "loss": 5.1125,
      "mean_token_accuracy": 0.18789087980985641,
      "num_tokens": 19742926.0,
      "step": 10725
    },
    {
      "entropy": 5.246692800521851,
      "epoch": 0.8962579351820915,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004925844940660928,
      "loss": 5.1071,
      "mean_token_accuracy": 0.19859821647405623,
      "num_tokens": 19752488.0,
      "step": 10730
    },
    {
      "entropy": 5.338521957397461,
      "epoch": 0.8966755763448045,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004925769122440606,
      "loss": 5.2231,
      "mean_token_accuracy": 0.1855644553899765,
      "num_tokens": 19762705.0,
      "step": 10735
    },
    {
      "entropy": 5.165931701660156,
      "epoch": 0.8970932175075176,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004925693266130824,
      "loss": 5.0002,
      "mean_token_accuracy": 0.20249298065900803,
      "num_tokens": 19771156.0,
      "step": 10740
    },
    {
      "entropy": 5.201911497116089,
      "epoch": 0.8975108586702305,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004925617371732914,
      "loss": 5.0717,
      "mean_token_accuracy": 0.19346583634614944,
      "num_tokens": 19780788.0,
      "step": 10745
    },
    {
      "entropy": 5.1557775974273685,
      "epoch": 0.8979284998329435,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004925541439248203,
      "loss": 5.0653,
      "mean_token_accuracy": 0.20302052944898605,
      "num_tokens": 19790169.0,
      "step": 10750
    },
    {
      "entropy": 5.376807689666748,
      "epoch": 0.8983461409956566,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000492546546867802,
      "loss": 5.2677,
      "mean_token_accuracy": 0.18198804259300233,
      "num_tokens": 19800249.0,
      "step": 10755
    },
    {
      "entropy": 5.302973747253418,
      "epoch": 0.8987637821583695,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004925389460023696,
      "loss": 5.1285,
      "mean_token_accuracy": 0.19517812877893448,
      "num_tokens": 19809797.0,
      "step": 10760
    },
    {
      "entropy": 5.21311206817627,
      "epoch": 0.8991814233210825,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004925313413286559,
      "loss": 5.0076,
      "mean_token_accuracy": 0.1987553909420967,
      "num_tokens": 19818826.0,
      "step": 10765
    },
    {
      "entropy": 5.244959831237793,
      "epoch": 0.8995990644837956,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004925237328467945,
      "loss": 5.1714,
      "mean_token_accuracy": 0.19508686363697053,
      "num_tokens": 19827947.0,
      "step": 10770
    },
    {
      "entropy": 5.292745637893677,
      "epoch": 0.9000167056465085,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004925161205569182,
      "loss": 5.1936,
      "mean_token_accuracy": 0.1910693258047104,
      "num_tokens": 19835734.0,
      "step": 10775
    },
    {
      "entropy": 5.3892723560333256,
      "epoch": 0.9004343468092215,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004925085044591604,
      "loss": 5.2883,
      "mean_token_accuracy": 0.18035610169172286,
      "num_tokens": 19845175.0,
      "step": 10780
    },
    {
      "entropy": 5.320371389389038,
      "epoch": 0.9008519879719346,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004925008845536544,
      "loss": 5.1323,
      "mean_token_accuracy": 0.19500856399536132,
      "num_tokens": 19853684.0,
      "step": 10785
    },
    {
      "entropy": 5.27652587890625,
      "epoch": 0.9012696291346475,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004924932608405336,
      "loss": 5.111,
      "mean_token_accuracy": 0.18882739543914795,
      "num_tokens": 19862813.0,
      "step": 10790
    },
    {
      "entropy": 5.276856994628906,
      "epoch": 0.9016872702973605,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004924856333199315,
      "loss": 5.1608,
      "mean_token_accuracy": 0.19190745502710344,
      "num_tokens": 19871545.0,
      "step": 10795
    },
    {
      "entropy": 5.249938869476319,
      "epoch": 0.9021049114600735,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004924780019919816,
      "loss": 5.1156,
      "mean_token_accuracy": 0.19244439005851746,
      "num_tokens": 19880060.0,
      "step": 10800
    },
    {
      "entropy": 5.247124195098877,
      "epoch": 0.9025225526227865,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004924703668568174,
      "loss": 5.1383,
      "mean_token_accuracy": 0.19215411245822905,
      "num_tokens": 19889151.0,
      "step": 10805
    },
    {
      "entropy": 5.318039798736573,
      "epoch": 0.9029401937854995,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004924627279145728,
      "loss": 5.1116,
      "mean_token_accuracy": 0.1943031892180443,
      "num_tokens": 19898087.0,
      "step": 10810
    },
    {
      "entropy": 5.221319770812988,
      "epoch": 0.9033578349482125,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004924550851653814,
      "loss": 5.1011,
      "mean_token_accuracy": 0.19722747504711152,
      "num_tokens": 19906196.0,
      "step": 10815
    },
    {
      "entropy": 5.267580986022949,
      "epoch": 0.9037754761109255,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004924474386093769,
      "loss": 5.1418,
      "mean_token_accuracy": 0.19102976620197296,
      "num_tokens": 19915837.0,
      "step": 10820
    },
    {
      "entropy": 5.225900363922119,
      "epoch": 0.9041931172736385,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004924397882466933,
      "loss": 5.1205,
      "mean_token_accuracy": 0.19227253943681716,
      "num_tokens": 19926012.0,
      "step": 10825
    },
    {
      "entropy": 5.231099605560303,
      "epoch": 0.9046107584363515,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004924321340774644,
      "loss": 5.0847,
      "mean_token_accuracy": 0.20060573220252992,
      "num_tokens": 19935987.0,
      "step": 10830
    },
    {
      "entropy": 5.2429530143737795,
      "epoch": 0.9050283995990644,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004924244761018243,
      "loss": 5.1254,
      "mean_token_accuracy": 0.1855889990925789,
      "num_tokens": 19945865.0,
      "step": 10835
    },
    {
      "entropy": 5.3116230964660645,
      "epoch": 0.9054460407617775,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000492416814319907,
      "loss": 5.1968,
      "mean_token_accuracy": 0.19410697221755982,
      "num_tokens": 19954824.0,
      "step": 10840
    },
    {
      "entropy": 5.253217697143555,
      "epoch": 0.9058636819244905,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004924091487318468,
      "loss": 5.086,
      "mean_token_accuracy": 0.1899289980530739,
      "num_tokens": 19964740.0,
      "step": 10845
    },
    {
      "entropy": 5.1915442943573,
      "epoch": 0.9062813230872034,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004924014793377776,
      "loss": 5.0781,
      "mean_token_accuracy": 0.1955967515707016,
      "num_tokens": 19974310.0,
      "step": 10850
    },
    {
      "entropy": 5.213424491882324,
      "epoch": 0.9066989642499165,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004923938061378339,
      "loss": 4.9736,
      "mean_token_accuracy": 0.20247429311275483,
      "num_tokens": 19982943.0,
      "step": 10855
    },
    {
      "entropy": 5.219340658187866,
      "epoch": 0.9071166054126295,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004923861291321499,
      "loss": 5.0949,
      "mean_token_accuracy": 0.19521151930093766,
      "num_tokens": 19991971.0,
      "step": 10860
    },
    {
      "entropy": 5.314491605758667,
      "epoch": 0.9075342465753424,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004923784483208601,
      "loss": 5.1636,
      "mean_token_accuracy": 0.18518495410680771,
      "num_tokens": 20001491.0,
      "step": 10865
    },
    {
      "entropy": 5.230543041229248,
      "epoch": 0.9079518877380555,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004923707637040989,
      "loss": 5.0426,
      "mean_token_accuracy": 0.19985975772142411,
      "num_tokens": 20010605.0,
      "step": 10870
    },
    {
      "entropy": 5.070496845245361,
      "epoch": 0.9083695289007685,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004923630752820008,
      "loss": 4.9238,
      "mean_token_accuracy": 0.2020262971520424,
      "num_tokens": 20019291.0,
      "step": 10875
    },
    {
      "entropy": 5.1730609893798825,
      "epoch": 0.9087871700634814,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004923553830547004,
      "loss": 5.2154,
      "mean_token_accuracy": 0.18565003871917723,
      "num_tokens": 20029048.0,
      "step": 10880
    },
    {
      "entropy": 5.346516752243042,
      "epoch": 0.9092048112261945,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004923476870223324,
      "loss": 5.1689,
      "mean_token_accuracy": 0.1868954837322235,
      "num_tokens": 20037878.0,
      "step": 10885
    },
    {
      "entropy": 5.2172050952911375,
      "epoch": 0.9096224523889075,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004923399871850315,
      "loss": 5.0856,
      "mean_token_accuracy": 0.19728281050920488,
      "num_tokens": 20046662.0,
      "step": 10890
    },
    {
      "entropy": 5.181592226028442,
      "epoch": 0.9100400935516204,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004923322835429326,
      "loss": 5.0638,
      "mean_token_accuracy": 0.19611624032258987,
      "num_tokens": 20055329.0,
      "step": 10895
    },
    {
      "entropy": 5.161191701889038,
      "epoch": 0.9104577347143334,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004923245760961704,
      "loss": 4.9831,
      "mean_token_accuracy": 0.19979818016290665,
      "num_tokens": 20064024.0,
      "step": 10900
    },
    {
      "entropy": 5.310274457931518,
      "epoch": 0.9108753758770465,
      "grad_norm": 1.5,
      "learning_rate": 0.0004923168648448799,
      "loss": 5.1841,
      "mean_token_accuracy": 0.18617172092199324,
      "num_tokens": 20073022.0,
      "step": 10905
    },
    {
      "entropy": 5.323524713516235,
      "epoch": 0.9112930170397594,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004923091497891961,
      "loss": 5.1462,
      "mean_token_accuracy": 0.18911670297384262,
      "num_tokens": 20081495.0,
      "step": 10910
    },
    {
      "entropy": 5.28730092048645,
      "epoch": 0.9117106582024724,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004923014309292541,
      "loss": 5.0904,
      "mean_token_accuracy": 0.1986383929848671,
      "num_tokens": 20090545.0,
      "step": 10915
    },
    {
      "entropy": 5.27257604598999,
      "epoch": 0.9121282993651855,
      "grad_norm": 1.5703125,
      "learning_rate": 0.000492293708265189,
      "loss": 5.1651,
      "mean_token_accuracy": 0.18962029218673707,
      "num_tokens": 20100236.0,
      "step": 10920
    },
    {
      "entropy": 5.223916673660279,
      "epoch": 0.9125459405278984,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004922859817971359,
      "loss": 5.0964,
      "mean_token_accuracy": 0.19232667833566666,
      "num_tokens": 20109522.0,
      "step": 10925
    },
    {
      "entropy": 5.23089337348938,
      "epoch": 0.9129635816906114,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004922782515252302,
      "loss": 5.0412,
      "mean_token_accuracy": 0.19769183546304703,
      "num_tokens": 20118248.0,
      "step": 10930
    },
    {
      "entropy": 5.151071214675904,
      "epoch": 0.9133812228533245,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004922705174496072,
      "loss": 5.1048,
      "mean_token_accuracy": 0.19247053414583207,
      "num_tokens": 20127096.0,
      "step": 10935
    },
    {
      "entropy": 5.231598663330078,
      "epoch": 0.9137988640160374,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004922627795704022,
      "loss": 5.0924,
      "mean_token_accuracy": 0.18838803470134735,
      "num_tokens": 20136815.0,
      "step": 10940
    },
    {
      "entropy": 5.260688161849975,
      "epoch": 0.9142165051787504,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004922550378877507,
      "loss": 5.0907,
      "mean_token_accuracy": 0.2012535959482193,
      "num_tokens": 20145455.0,
      "step": 10945
    },
    {
      "entropy": 5.281193494796753,
      "epoch": 0.9146341463414634,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004922472924017882,
      "loss": 5.1899,
      "mean_token_accuracy": 0.19064459353685378,
      "num_tokens": 20156139.0,
      "step": 10950
    },
    {
      "entropy": 5.374291896820068,
      "epoch": 0.9150517875041764,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004922395431126506,
      "loss": 5.1562,
      "mean_token_accuracy": 0.19103390276432036,
      "num_tokens": 20166266.0,
      "step": 10955
    },
    {
      "entropy": 5.356672620773315,
      "epoch": 0.9154694286668894,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004922317900204732,
      "loss": 5.214,
      "mean_token_accuracy": 0.18659820705652236,
      "num_tokens": 20175005.0,
      "step": 10960
    },
    {
      "entropy": 5.187180995941162,
      "epoch": 0.9158870698296024,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004922240331253916,
      "loss": 5.0272,
      "mean_token_accuracy": 0.20363794565200805,
      "num_tokens": 20183488.0,
      "step": 10965
    },
    {
      "entropy": 5.226777267456055,
      "epoch": 0.9163047109923154,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004922162724275421,
      "loss": 5.1089,
      "mean_token_accuracy": 0.19940198659896852,
      "num_tokens": 20193724.0,
      "step": 10970
    },
    {
      "entropy": 5.279091882705688,
      "epoch": 0.9167223521550284,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004922085079270602,
      "loss": 5.2065,
      "mean_token_accuracy": 0.19198762327432634,
      "num_tokens": 20203733.0,
      "step": 10975
    },
    {
      "entropy": 5.406766319274903,
      "epoch": 0.9171399933177414,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004922007396240819,
      "loss": 5.1745,
      "mean_token_accuracy": 0.19407681077718736,
      "num_tokens": 20213448.0,
      "step": 10980
    },
    {
      "entropy": 5.275003099441529,
      "epoch": 0.9175576344804544,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004921929675187433,
      "loss": 5.0912,
      "mean_token_accuracy": 0.1954730987548828,
      "num_tokens": 20222837.0,
      "step": 10985
    },
    {
      "entropy": 5.275844430923462,
      "epoch": 0.9179752756431674,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004921851916111802,
      "loss": 5.2749,
      "mean_token_accuracy": 0.18780399411916732,
      "num_tokens": 20232056.0,
      "step": 10990
    },
    {
      "entropy": 5.3480713844299315,
      "epoch": 0.9183929168058804,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004921774119015289,
      "loss": 5.1913,
      "mean_token_accuracy": 0.1876297414302826,
      "num_tokens": 20241052.0,
      "step": 10995
    },
    {
      "entropy": 5.357371425628662,
      "epoch": 0.9188105579685933,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004921696283899257,
      "loss": 5.242,
      "mean_token_accuracy": 0.18337310701608658,
      "num_tokens": 20251015.0,
      "step": 11000
    },
    {
      "entropy": 5.26271686553955,
      "epoch": 0.9192281991313064,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004921618410765067,
      "loss": 5.0737,
      "mean_token_accuracy": 0.19415029585361482,
      "num_tokens": 20259388.0,
      "step": 11005
    },
    {
      "entropy": 5.304884433746338,
      "epoch": 0.9196458402940194,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004921540499614083,
      "loss": 5.0629,
      "mean_token_accuracy": 0.19187703132629394,
      "num_tokens": 20268146.0,
      "step": 11010
    },
    {
      "entropy": 5.202958774566651,
      "epoch": 0.9200634814567323,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004921462550447669,
      "loss": 5.038,
      "mean_token_accuracy": 0.19929284155368804,
      "num_tokens": 20277559.0,
      "step": 11015
    },
    {
      "entropy": 5.219947385787964,
      "epoch": 0.9204811226194454,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004921384563267189,
      "loss": 5.107,
      "mean_token_accuracy": 0.1902410179376602,
      "num_tokens": 20285661.0,
      "step": 11020
    },
    {
      "entropy": 5.273578119277954,
      "epoch": 0.9208987637821584,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004921306538074008,
      "loss": 5.1763,
      "mean_token_accuracy": 0.19261985421180725,
      "num_tokens": 20294636.0,
      "step": 11025
    },
    {
      "entropy": 5.291933536529541,
      "epoch": 0.9213164049448713,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004921228474869493,
      "loss": 5.0787,
      "mean_token_accuracy": 0.20037764608860015,
      "num_tokens": 20302372.0,
      "step": 11030
    },
    {
      "entropy": 5.2915506839752195,
      "epoch": 0.9217340461075844,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000492115037365501,
      "loss": 5.0953,
      "mean_token_accuracy": 0.19168005436658858,
      "num_tokens": 20311974.0,
      "step": 11035
    },
    {
      "entropy": 5.228907108306885,
      "epoch": 0.9221516872702974,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004921072234431925,
      "loss": 5.1177,
      "mean_token_accuracy": 0.19093839079141617,
      "num_tokens": 20321720.0,
      "step": 11040
    },
    {
      "entropy": 5.263603019714355,
      "epoch": 0.9225693284330103,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000492099405720161,
      "loss": 5.1695,
      "mean_token_accuracy": 0.18371784090995788,
      "num_tokens": 20331324.0,
      "step": 11045
    },
    {
      "entropy": 5.381476211547851,
      "epoch": 0.9229869695957233,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004920915841965429,
      "loss": 5.2627,
      "mean_token_accuracy": 0.187418133020401,
      "num_tokens": 20340921.0,
      "step": 11050
    },
    {
      "entropy": 5.36841230392456,
      "epoch": 0.9234046107584364,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004920837588724753,
      "loss": 5.1272,
      "mean_token_accuracy": 0.1968556046485901,
      "num_tokens": 20349871.0,
      "step": 11055
    },
    {
      "entropy": 5.130883312225341,
      "epoch": 0.9238222519211493,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004920759297480953,
      "loss": 4.9175,
      "mean_token_accuracy": 0.1999177008867264,
      "num_tokens": 20359171.0,
      "step": 11060
    },
    {
      "entropy": 5.19677243232727,
      "epoch": 0.9242398930838623,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004920680968235397,
      "loss": 5.1201,
      "mean_token_accuracy": 0.19957246780395507,
      "num_tokens": 20367779.0,
      "step": 11065
    },
    {
      "entropy": 5.197241687774659,
      "epoch": 0.9246575342465754,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004920602600989459,
      "loss": 5.0436,
      "mean_token_accuracy": 0.2011151969432831,
      "num_tokens": 20377568.0,
      "step": 11070
    },
    {
      "entropy": 5.268267250061035,
      "epoch": 0.9250751754092883,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004920524195744509,
      "loss": 5.1783,
      "mean_token_accuracy": 0.18935236036777497,
      "num_tokens": 20385915.0,
      "step": 11075
    },
    {
      "entropy": 5.224002456665039,
      "epoch": 0.9254928165720013,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004920445752501921,
      "loss": 5.0542,
      "mean_token_accuracy": 0.19511025846004487,
      "num_tokens": 20394877.0,
      "step": 11080
    },
    {
      "entropy": 5.217405605316162,
      "epoch": 0.9259104577347144,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004920367271263066,
      "loss": 5.0424,
      "mean_token_accuracy": 0.19762918800115586,
      "num_tokens": 20403741.0,
      "step": 11085
    },
    {
      "entropy": 5.19131031036377,
      "epoch": 0.9263280988974273,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004920288752029321,
      "loss": 5.0818,
      "mean_token_accuracy": 0.19510875791311263,
      "num_tokens": 20412702.0,
      "step": 11090
    },
    {
      "entropy": 5.240503454208374,
      "epoch": 0.9267457400601403,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004920210194802059,
      "loss": 5.1383,
      "mean_token_accuracy": 0.19578883051872253,
      "num_tokens": 20421776.0,
      "step": 11095
    },
    {
      "entropy": 5.323799133300781,
      "epoch": 0.9271633812228534,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004920131599582654,
      "loss": 5.2297,
      "mean_token_accuracy": 0.17944206148386002,
      "num_tokens": 20432204.0,
      "step": 11100
    },
    {
      "entropy": 5.284155082702637,
      "epoch": 0.9275810223855663,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004920052966372485,
      "loss": 5.118,
      "mean_token_accuracy": 0.1872738242149353,
      "num_tokens": 20441344.0,
      "step": 11105
    },
    {
      "entropy": 5.226302289962769,
      "epoch": 0.9279986635482793,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004919974295172926,
      "loss": 5.0325,
      "mean_token_accuracy": 0.19532300680875778,
      "num_tokens": 20449800.0,
      "step": 11110
    },
    {
      "entropy": 5.212497711181641,
      "epoch": 0.9284163047109923,
      "grad_norm": 1.5,
      "learning_rate": 0.0004919895585985353,
      "loss": 4.982,
      "mean_token_accuracy": 0.20732794255018233,
      "num_tokens": 20458326.0,
      "step": 11115
    },
    {
      "entropy": 5.2523040771484375,
      "epoch": 0.9288339458737053,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004919816838811149,
      "loss": 5.249,
      "mean_token_accuracy": 0.1881237119436264,
      "num_tokens": 20468050.0,
      "step": 11120
    },
    {
      "entropy": 5.238019037246704,
      "epoch": 0.9292515870364183,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004919738053651687,
      "loss": 5.0028,
      "mean_token_accuracy": 0.1985087901353836,
      "num_tokens": 20477046.0,
      "step": 11125
    },
    {
      "entropy": 5.230250692367553,
      "epoch": 0.9296692281991313,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000491965923050835,
      "loss": 5.1172,
      "mean_token_accuracy": 0.19466379582881926,
      "num_tokens": 20486228.0,
      "step": 11130
    },
    {
      "entropy": 5.268046951293945,
      "epoch": 0.9300868693618443,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004919580369382516,
      "loss": 5.1408,
      "mean_token_accuracy": 0.18756549209356307,
      "num_tokens": 20494764.0,
      "step": 11135
    },
    {
      "entropy": 5.255468225479126,
      "epoch": 0.9305045105245573,
      "grad_norm": 1.5,
      "learning_rate": 0.0004919501470275567,
      "loss": 5.0643,
      "mean_token_accuracy": 0.19451435953378676,
      "num_tokens": 20502805.0,
      "step": 11140
    },
    {
      "entropy": 5.218306303024292,
      "epoch": 0.9309221516872703,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004919422533188882,
      "loss": 5.1011,
      "mean_token_accuracy": 0.19301692098379136,
      "num_tokens": 20512322.0,
      "step": 11145
    },
    {
      "entropy": 5.207473564147949,
      "epoch": 0.9313397928499833,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004919343558123845,
      "loss": 5.0982,
      "mean_token_accuracy": 0.19467852115631104,
      "num_tokens": 20520680.0,
      "step": 11150
    },
    {
      "entropy": 5.199759769439697,
      "epoch": 0.9317574340126963,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004919264545081838,
      "loss": 4.9695,
      "mean_token_accuracy": 0.20823939740657807,
      "num_tokens": 20528864.0,
      "step": 11155
    },
    {
      "entropy": 5.223690223693848,
      "epoch": 0.9321750751754093,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004919185494064244,
      "loss": 5.0881,
      "mean_token_accuracy": 0.18788518607616425,
      "num_tokens": 20538195.0,
      "step": 11160
    },
    {
      "entropy": 5.2309746742248535,
      "epoch": 0.9325927163381222,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004919106405072446,
      "loss": 5.1309,
      "mean_token_accuracy": 0.19108415991067887,
      "num_tokens": 20547312.0,
      "step": 11165
    },
    {
      "entropy": 5.315044403076172,
      "epoch": 0.9330103575008353,
      "grad_norm": 1.28125,
      "learning_rate": 0.000491902727810783,
      "loss": 5.1638,
      "mean_token_accuracy": 0.1868708312511444,
      "num_tokens": 20557184.0,
      "step": 11170
    },
    {
      "entropy": 5.237520027160644,
      "epoch": 0.9334279986635483,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004918948113171781,
      "loss": 5.0959,
      "mean_token_accuracy": 0.19299858808517456,
      "num_tokens": 20565819.0,
      "step": 11175
    },
    {
      "entropy": 5.222719335556031,
      "epoch": 0.9338456398262612,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004918868910265684,
      "loss": 5.1486,
      "mean_token_accuracy": 0.19817730635404587,
      "num_tokens": 20574761.0,
      "step": 11180
    },
    {
      "entropy": 5.358972263336182,
      "epoch": 0.9342632809889743,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004918789669390927,
      "loss": 5.2016,
      "mean_token_accuracy": 0.1851782202720642,
      "num_tokens": 20584003.0,
      "step": 11185
    },
    {
      "entropy": 5.277846145629883,
      "epoch": 0.9346809221516873,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004918710390548895,
      "loss": 5.0554,
      "mean_token_accuracy": 0.19363753199577333,
      "num_tokens": 20593123.0,
      "step": 11190
    },
    {
      "entropy": 5.232641267776489,
      "epoch": 0.9350985633144002,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004918631073740977,
      "loss": 5.1326,
      "mean_token_accuracy": 0.19118748903274535,
      "num_tokens": 20602582.0,
      "step": 11195
    },
    {
      "entropy": 5.21953330039978,
      "epoch": 0.9355162044771133,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004918551718968562,
      "loss": 5.046,
      "mean_token_accuracy": 0.19664665013551713,
      "num_tokens": 20611681.0,
      "step": 11200
    },
    {
      "entropy": 5.217195749282837,
      "epoch": 0.9359338456398263,
      "grad_norm": 1.328125,
      "learning_rate": 0.000491847232623304,
      "loss": 5.1133,
      "mean_token_accuracy": 0.19649536311626434,
      "num_tokens": 20620671.0,
      "step": 11205
    },
    {
      "entropy": 5.214206314086914,
      "epoch": 0.9363514868025392,
      "grad_norm": 1.5,
      "learning_rate": 0.0004918392895535798,
      "loss": 5.0722,
      "mean_token_accuracy": 0.19406209290027618,
      "num_tokens": 20629986.0,
      "step": 11210
    },
    {
      "entropy": 5.211397266387939,
      "epoch": 0.9367691279652522,
      "grad_norm": 1.25,
      "learning_rate": 0.000491831342687823,
      "loss": 5.0555,
      "mean_token_accuracy": 0.1930464804172516,
      "num_tokens": 20639274.0,
      "step": 11215
    },
    {
      "entropy": 5.23776159286499,
      "epoch": 0.9371867691279653,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004918233920261723,
      "loss": 5.0931,
      "mean_token_accuracy": 0.19755666106939315,
      "num_tokens": 20647753.0,
      "step": 11220
    },
    {
      "entropy": 5.136107063293457,
      "epoch": 0.9376044102906782,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004918154375687673,
      "loss": 4.9853,
      "mean_token_accuracy": 0.20235325396060944,
      "num_tokens": 20656226.0,
      "step": 11225
    },
    {
      "entropy": 5.152089643478393,
      "epoch": 0.9380220514533912,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004918074793157472,
      "loss": 5.0115,
      "mean_token_accuracy": 0.19989002794027327,
      "num_tokens": 20664682.0,
      "step": 11230
    },
    {
      "entropy": 5.283812999725342,
      "epoch": 0.9384396926161043,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004917995172672511,
      "loss": 5.1058,
      "mean_token_accuracy": 0.19565124213695526,
      "num_tokens": 20672909.0,
      "step": 11235
    },
    {
      "entropy": 5.1583353042602536,
      "epoch": 0.9388573337788172,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004917915514234184,
      "loss": 4.944,
      "mean_token_accuracy": 0.20490019023418427,
      "num_tokens": 20681316.0,
      "step": 11240
    },
    {
      "entropy": 5.2468834400177,
      "epoch": 0.9392749749415302,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004917835817843887,
      "loss": 5.0597,
      "mean_token_accuracy": 0.18870282918214798,
      "num_tokens": 20690360.0,
      "step": 11245
    },
    {
      "entropy": 5.291747283935547,
      "epoch": 0.9396926161042433,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004917756083503014,
      "loss": 5.1129,
      "mean_token_accuracy": 0.1873975306749344,
      "num_tokens": 20700398.0,
      "step": 11250
    },
    {
      "entropy": 5.305414819717408,
      "epoch": 0.9401102572669562,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004917676311212962,
      "loss": 5.213,
      "mean_token_accuracy": 0.18390239477157594,
      "num_tokens": 20709705.0,
      "step": 11255
    },
    {
      "entropy": 5.149075746536255,
      "epoch": 0.9405278984296692,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004917596500975127,
      "loss": 5.0045,
      "mean_token_accuracy": 0.20366675704717635,
      "num_tokens": 20717774.0,
      "step": 11260
    },
    {
      "entropy": 5.275296211242676,
      "epoch": 0.9409455395923823,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004917516652790907,
      "loss": 5.1619,
      "mean_token_accuracy": 0.19286663234233856,
      "num_tokens": 20725966.0,
      "step": 11265
    },
    {
      "entropy": 5.298273181915283,
      "epoch": 0.9413631807550952,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004917436766661697,
      "loss": 5.0607,
      "mean_token_accuracy": 0.19599981606006622,
      "num_tokens": 20735254.0,
      "step": 11270
    },
    {
      "entropy": 5.199138069152832,
      "epoch": 0.9417808219178082,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00049173568425889,
      "loss": 4.9829,
      "mean_token_accuracy": 0.2004885643720627,
      "num_tokens": 20743331.0,
      "step": 11275
    },
    {
      "entropy": 5.22732195854187,
      "epoch": 0.9421984630805212,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004917276880573911,
      "loss": 5.0976,
      "mean_token_accuracy": 0.18877559006214142,
      "num_tokens": 20753966.0,
      "step": 11280
    },
    {
      "entropy": 5.21967921257019,
      "epoch": 0.9426161042432342,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004917196880618132,
      "loss": 5.0479,
      "mean_token_accuracy": 0.19432479292154312,
      "num_tokens": 20762608.0,
      "step": 11285
    },
    {
      "entropy": 5.121931266784668,
      "epoch": 0.9430337454059472,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004917116842722962,
      "loss": 4.982,
      "mean_token_accuracy": 0.19571133255958556,
      "num_tokens": 20771194.0,
      "step": 11290
    },
    {
      "entropy": 5.24848256111145,
      "epoch": 0.9434513865686602,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004917036766889804,
      "loss": 5.0855,
      "mean_token_accuracy": 0.20109958201646805,
      "num_tokens": 20779700.0,
      "step": 11295
    },
    {
      "entropy": 5.303081321716308,
      "epoch": 0.9438690277313732,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004916956653120058,
      "loss": 5.2234,
      "mean_token_accuracy": 0.18504958152770995,
      "num_tokens": 20788117.0,
      "step": 11300
    },
    {
      "entropy": 5.300807189941406,
      "epoch": 0.9442866688940862,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004916876501415129,
      "loss": 5.187,
      "mean_token_accuracy": 0.18734770715236665,
      "num_tokens": 20798041.0,
      "step": 11305
    },
    {
      "entropy": 5.290560960769653,
      "epoch": 0.9447043100567992,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004916796311776417,
      "loss": 5.1201,
      "mean_token_accuracy": 0.1854579895734787,
      "num_tokens": 20807423.0,
      "step": 11310
    },
    {
      "entropy": 5.211841869354248,
      "epoch": 0.9451219512195121,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004916716084205328,
      "loss": 5.0247,
      "mean_token_accuracy": 0.19960433691740037,
      "num_tokens": 20816727.0,
      "step": 11315
    },
    {
      "entropy": 5.188795804977417,
      "epoch": 0.9455395923822252,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004916635818703266,
      "loss": 5.036,
      "mean_token_accuracy": 0.19338960945606232,
      "num_tokens": 20824855.0,
      "step": 11320
    },
    {
      "entropy": 5.127743721008301,
      "epoch": 0.9459572335449382,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004916555515271635,
      "loss": 5.0579,
      "mean_token_accuracy": 0.20120692402124404,
      "num_tokens": 20833096.0,
      "step": 11325
    },
    {
      "entropy": 5.17639479637146,
      "epoch": 0.9463748747076511,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004916475173911844,
      "loss": 5.0202,
      "mean_token_accuracy": 0.19952136278152466,
      "num_tokens": 20842398.0,
      "step": 11330
    },
    {
      "entropy": 5.366609001159668,
      "epoch": 0.9467925158703642,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004916394794625296,
      "loss": 5.2234,
      "mean_token_accuracy": 0.18625444620847703,
      "num_tokens": 20851745.0,
      "step": 11335
    },
    {
      "entropy": 5.396459102630615,
      "epoch": 0.9472101570330772,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004916314377413399,
      "loss": 5.1447,
      "mean_token_accuracy": 0.19049628227949142,
      "num_tokens": 20860935.0,
      "step": 11340
    },
    {
      "entropy": 5.222161865234375,
      "epoch": 0.9476277981957901,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004916233922277562,
      "loss": 5.1238,
      "mean_token_accuracy": 0.19106235802173616,
      "num_tokens": 20871088.0,
      "step": 11345
    },
    {
      "entropy": 5.221987724304199,
      "epoch": 0.9480454393585032,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004916153429219192,
      "loss": 5.1677,
      "mean_token_accuracy": 0.19065823405981064,
      "num_tokens": 20879313.0,
      "step": 11350
    },
    {
      "entropy": 5.261417293548584,
      "epoch": 0.9484630805212162,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004916072898239699,
      "loss": 5.1134,
      "mean_token_accuracy": 0.18942350447177886,
      "num_tokens": 20888392.0,
      "step": 11355
    },
    {
      "entropy": 5.280555963516235,
      "epoch": 0.9488807216839291,
      "grad_norm": 1.375,
      "learning_rate": 0.0004915992329340493,
      "loss": 5.2131,
      "mean_token_accuracy": 0.18295790404081344,
      "num_tokens": 20897456.0,
      "step": 11360
    },
    {
      "entropy": 5.283572006225586,
      "epoch": 0.9492983628466422,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004915911722522984,
      "loss": 5.1186,
      "mean_token_accuracy": 0.1935260385274887,
      "num_tokens": 20906621.0,
      "step": 11365
    },
    {
      "entropy": 5.3734008312225345,
      "epoch": 0.9497160040093552,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004915831077788583,
      "loss": 5.2827,
      "mean_token_accuracy": 0.18720433712005616,
      "num_tokens": 20916929.0,
      "step": 11370
    },
    {
      "entropy": 5.254325675964355,
      "epoch": 0.9501336451720681,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004915750395138702,
      "loss": 5.1134,
      "mean_token_accuracy": 0.1938747614622116,
      "num_tokens": 20926028.0,
      "step": 11375
    },
    {
      "entropy": 5.292331743240356,
      "epoch": 0.9505512863347811,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004915669674574754,
      "loss": 5.2062,
      "mean_token_accuracy": 0.18499393463134767,
      "num_tokens": 20935516.0,
      "step": 11380
    },
    {
      "entropy": 5.231068706512451,
      "epoch": 0.9509689274974942,
      "grad_norm": 1.375,
      "learning_rate": 0.0004915588916098151,
      "loss": 5.0001,
      "mean_token_accuracy": 0.20366467386484147,
      "num_tokens": 20944100.0,
      "step": 11385
    },
    {
      "entropy": 5.283082485198975,
      "epoch": 0.9513865686602071,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004915508119710307,
      "loss": 5.1374,
      "mean_token_accuracy": 0.1924660846590996,
      "num_tokens": 20953733.0,
      "step": 11390
    },
    {
      "entropy": 5.313457727432251,
      "epoch": 0.9518042098229201,
      "grad_norm": 1.5,
      "learning_rate": 0.0004915427285412636,
      "loss": 5.1428,
      "mean_token_accuracy": 0.1839990049600601,
      "num_tokens": 20962413.0,
      "step": 11395
    },
    {
      "entropy": 5.165603971481323,
      "epoch": 0.9522218509856332,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004915346413206555,
      "loss": 5.0033,
      "mean_token_accuracy": 0.2058175429701805,
      "num_tokens": 20971172.0,
      "step": 11400
    },
    {
      "entropy": 5.1996723175048825,
      "epoch": 0.9526394921483461,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004915265503093478,
      "loss": 5.0295,
      "mean_token_accuracy": 0.1958525598049164,
      "num_tokens": 20979450.0,
      "step": 11405
    },
    {
      "entropy": 5.221041059494018,
      "epoch": 0.9530571333110591,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004915184555074824,
      "loss": 5.1084,
      "mean_token_accuracy": 0.18944413661956788,
      "num_tokens": 20988518.0,
      "step": 11410
    },
    {
      "entropy": 5.393524646759033,
      "epoch": 0.9534747744737722,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004915103569152005,
      "loss": 5.2763,
      "mean_token_accuracy": 0.18145391196012498,
      "num_tokens": 20998721.0,
      "step": 11415
    },
    {
      "entropy": 5.328356599807739,
      "epoch": 0.9538924156364851,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004915022545326444,
      "loss": 5.1397,
      "mean_token_accuracy": 0.19815013706684112,
      "num_tokens": 21008007.0,
      "step": 11420
    },
    {
      "entropy": 5.204979085922242,
      "epoch": 0.9543100567991981,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004914941483599558,
      "loss": 5.2337,
      "mean_token_accuracy": 0.1896138459444046,
      "num_tokens": 21017199.0,
      "step": 11425
    },
    {
      "entropy": 5.110478115081787,
      "epoch": 0.9547276979619111,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004914860383972763,
      "loss": 5.0113,
      "mean_token_accuracy": 0.19855717569589615,
      "num_tokens": 21025571.0,
      "step": 11430
    },
    {
      "entropy": 5.276226282119751,
      "epoch": 0.9551453391246241,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004914779246447483,
      "loss": 5.109,
      "mean_token_accuracy": 0.19669386893510818,
      "num_tokens": 21035040.0,
      "step": 11435
    },
    {
      "entropy": 5.2725029468536375,
      "epoch": 0.9555629802873371,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004914698071025136,
      "loss": 5.148,
      "mean_token_accuracy": 0.19776418507099153,
      "num_tokens": 21044018.0,
      "step": 11440
    },
    {
      "entropy": 5.239176464080811,
      "epoch": 0.9559806214500501,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004914616857707143,
      "loss": 5.0453,
      "mean_token_accuracy": 0.20217226445674896,
      "num_tokens": 21052489.0,
      "step": 11445
    },
    {
      "entropy": 5.247550439834595,
      "epoch": 0.9563982626127631,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004914535606494927,
      "loss": 5.1228,
      "mean_token_accuracy": 0.19844011813402176,
      "num_tokens": 21061153.0,
      "step": 11450
    },
    {
      "entropy": 5.20886754989624,
      "epoch": 0.9568159037754761,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000491445431738991,
      "loss": 5.0046,
      "mean_token_accuracy": 0.1983945280313492,
      "num_tokens": 21069671.0,
      "step": 11455
    },
    {
      "entropy": 5.355167055130005,
      "epoch": 0.9572335449381891,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004914372990393515,
      "loss": 5.3114,
      "mean_token_accuracy": 0.1806883230805397,
      "num_tokens": 21079339.0,
      "step": 11460
    },
    {
      "entropy": 5.21011438369751,
      "epoch": 0.957651186100902,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004914291625507164,
      "loss": 5.0124,
      "mean_token_accuracy": 0.19901425242424012,
      "num_tokens": 21087868.0,
      "step": 11465
    },
    {
      "entropy": 5.24385027885437,
      "epoch": 0.9580688272636151,
      "grad_norm": 1.5,
      "learning_rate": 0.0004914210222732284,
      "loss": 5.1263,
      "mean_token_accuracy": 0.19061318337917327,
      "num_tokens": 21096258.0,
      "step": 11470
    },
    {
      "entropy": 5.2431365013122555,
      "epoch": 0.9584864684263281,
      "grad_norm": 1.25,
      "learning_rate": 0.0004914128782070299,
      "loss": 5.1402,
      "mean_token_accuracy": 0.1892012134194374,
      "num_tokens": 21104504.0,
      "step": 11475
    },
    {
      "entropy": 5.272426176071167,
      "epoch": 0.958904109589041,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004914047303522636,
      "loss": 5.0665,
      "mean_token_accuracy": 0.1918003499507904,
      "num_tokens": 21114402.0,
      "step": 11480
    },
    {
      "entropy": 5.263548707962036,
      "epoch": 0.9593217507517541,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004913965787090718,
      "loss": 5.0871,
      "mean_token_accuracy": 0.19472704827785492,
      "num_tokens": 21123572.0,
      "step": 11485
    },
    {
      "entropy": 5.30646243095398,
      "epoch": 0.9597393919144671,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004913884232775974,
      "loss": 5.2875,
      "mean_token_accuracy": 0.1768502712249756,
      "num_tokens": 21132385.0,
      "step": 11490
    },
    {
      "entropy": 5.215727138519287,
      "epoch": 0.96015703307718,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004913802640579833,
      "loss": 5.0357,
      "mean_token_accuracy": 0.2048182725906372,
      "num_tokens": 21140383.0,
      "step": 11495
    },
    {
      "entropy": 5.317690134048462,
      "epoch": 0.9605746742398931,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004913721010503722,
      "loss": 5.1455,
      "mean_token_accuracy": 0.18924759477376937,
      "num_tokens": 21149793.0,
      "step": 11500
    },
    {
      "entropy": 5.245128440856933,
      "epoch": 0.9609923154026061,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000491363934254907,
      "loss": 5.1129,
      "mean_token_accuracy": 0.18947928845882417,
      "num_tokens": 21158955.0,
      "step": 11505
    },
    {
      "entropy": 5.297966146469117,
      "epoch": 0.961409956565319,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004913557636717307,
      "loss": 5.1163,
      "mean_token_accuracy": 0.19114254117012025,
      "num_tokens": 21167900.0,
      "step": 11510
    },
    {
      "entropy": 5.232564783096313,
      "epoch": 0.9618275977280321,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004913475893009864,
      "loss": 5.048,
      "mean_token_accuracy": 0.2020498752593994,
      "num_tokens": 21177243.0,
      "step": 11515
    },
    {
      "entropy": 5.295079469680786,
      "epoch": 0.9622452388907451,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000491339411142817,
      "loss": 5.1433,
      "mean_token_accuracy": 0.19128964692354203,
      "num_tokens": 21186118.0,
      "step": 11520
    },
    {
      "entropy": 5.2694587230682375,
      "epoch": 0.962662880053458,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004913312291973659,
      "loss": 5.1586,
      "mean_token_accuracy": 0.1919132202863693,
      "num_tokens": 21195032.0,
      "step": 11525
    },
    {
      "entropy": 5.265349340438843,
      "epoch": 0.963080521216171,
      "grad_norm": 1.375,
      "learning_rate": 0.0004913230434647762,
      "loss": 5.1359,
      "mean_token_accuracy": 0.19201079607009888,
      "num_tokens": 21205388.0,
      "step": 11530
    },
    {
      "entropy": 5.26227068901062,
      "epoch": 0.9634981623788841,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004913148539451912,
      "loss": 5.0336,
      "mean_token_accuracy": 0.19470599740743638,
      "num_tokens": 21215181.0,
      "step": 11535
    },
    {
      "entropy": 5.175901889801025,
      "epoch": 0.963915803541597,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004913066606387544,
      "loss": 4.9395,
      "mean_token_accuracy": 0.20754334032535554,
      "num_tokens": 21223668.0,
      "step": 11540
    },
    {
      "entropy": 5.155073022842407,
      "epoch": 0.96433344470431,
      "grad_norm": 1.296875,
      "learning_rate": 0.000491298463545609,
      "loss": 5.0732,
      "mean_token_accuracy": 0.19140748977661132,
      "num_tokens": 21232448.0,
      "step": 11545
    },
    {
      "entropy": 5.205829524993897,
      "epoch": 0.9647510858670231,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004912902626658988,
      "loss": 5.0482,
      "mean_token_accuracy": 0.1945975735783577,
      "num_tokens": 21240777.0,
      "step": 11550
    },
    {
      "entropy": 5.295857000350952,
      "epoch": 0.965168727029736,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004912820579997672,
      "loss": 5.1916,
      "mean_token_accuracy": 0.18431014865636824,
      "num_tokens": 21248803.0,
      "step": 11555
    },
    {
      "entropy": 5.294878196716309,
      "epoch": 0.965586368192449,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004912738495473577,
      "loss": 5.1345,
      "mean_token_accuracy": 0.18914187103509902,
      "num_tokens": 21257861.0,
      "step": 11560
    },
    {
      "entropy": 5.299514389038086,
      "epoch": 0.9660040093551621,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004912656373088142,
      "loss": 5.0434,
      "mean_token_accuracy": 0.1987417757511139,
      "num_tokens": 21267140.0,
      "step": 11565
    },
    {
      "entropy": 5.25201187133789,
      "epoch": 0.9664216505178751,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004912574212842804,
      "loss": 5.1448,
      "mean_token_accuracy": 0.19093319177627563,
      "num_tokens": 21276313.0,
      "step": 11570
    },
    {
      "entropy": 5.184623098373413,
      "epoch": 0.966839291680588,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004912492014739001,
      "loss": 5.0652,
      "mean_token_accuracy": 0.19548497200012208,
      "num_tokens": 21286156.0,
      "step": 11575
    },
    {
      "entropy": 5.2919511795043945,
      "epoch": 0.967256932843301,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004912409778778173,
      "loss": 5.1452,
      "mean_token_accuracy": 0.19334112703800202,
      "num_tokens": 21295420.0,
      "step": 11580
    },
    {
      "entropy": 5.342772626876831,
      "epoch": 0.9676745740060141,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004912327504961758,
      "loss": 5.3118,
      "mean_token_accuracy": 0.1796220377087593,
      "num_tokens": 21306292.0,
      "step": 11585
    },
    {
      "entropy": 5.290644598007202,
      "epoch": 0.968092215168727,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004912245193291198,
      "loss": 5.1474,
      "mean_token_accuracy": 0.19274698346853256,
      "num_tokens": 21315165.0,
      "step": 11590
    },
    {
      "entropy": 5.288678407669067,
      "epoch": 0.96850985633144,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004912162843767932,
      "loss": 5.1344,
      "mean_token_accuracy": 0.19493958055973054,
      "num_tokens": 21324608.0,
      "step": 11595
    },
    {
      "entropy": 5.299476957321167,
      "epoch": 0.9689274974941531,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004912080456393404,
      "loss": 5.0867,
      "mean_token_accuracy": 0.19237823337316512,
      "num_tokens": 21333609.0,
      "step": 11600
    },
    {
      "entropy": 5.286192846298218,
      "epoch": 0.969345138656866,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004911998031169056,
      "loss": 5.213,
      "mean_token_accuracy": 0.18236109167337416,
      "num_tokens": 21344397.0,
      "step": 11605
    },
    {
      "entropy": 5.3327747821807865,
      "epoch": 0.969762779819579,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004911915568096328,
      "loss": 5.2071,
      "mean_token_accuracy": 0.18246063590049744,
      "num_tokens": 21354111.0,
      "step": 11610
    },
    {
      "entropy": 5.159500312805176,
      "epoch": 0.9701804209822921,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004911833067176666,
      "loss": 4.987,
      "mean_token_accuracy": 0.19307583719491958,
      "num_tokens": 21362807.0,
      "step": 11615
    },
    {
      "entropy": 5.299811935424804,
      "epoch": 0.970598062145005,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004911750528411515,
      "loss": 5.1643,
      "mean_token_accuracy": 0.1926438644528389,
      "num_tokens": 21372609.0,
      "step": 11620
    },
    {
      "entropy": 5.238452053070068,
      "epoch": 0.971015703307718,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004911667951802318,
      "loss": 5.0852,
      "mean_token_accuracy": 0.1913050204515457,
      "num_tokens": 21381182.0,
      "step": 11625
    },
    {
      "entropy": 5.230159187316895,
      "epoch": 0.9714333444704311,
      "grad_norm": 1.25,
      "learning_rate": 0.0004911585337350524,
      "loss": 5.087,
      "mean_token_accuracy": 0.19937516301870345,
      "num_tokens": 21389930.0,
      "step": 11630
    },
    {
      "entropy": 5.3177858829498295,
      "epoch": 0.971850985633144,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004911502685057573,
      "loss": 5.15,
      "mean_token_accuracy": 0.1849372699856758,
      "num_tokens": 21399315.0,
      "step": 11635
    },
    {
      "entropy": 5.285046243667603,
      "epoch": 0.972268626795857,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004911419994924918,
      "loss": 5.1712,
      "mean_token_accuracy": 0.19047422409057618,
      "num_tokens": 21408857.0,
      "step": 11640
    },
    {
      "entropy": 5.2321692943573,
      "epoch": 0.97268626795857,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004911337266954004,
      "loss": 5.0708,
      "mean_token_accuracy": 0.1990445852279663,
      "num_tokens": 21417452.0,
      "step": 11645
    },
    {
      "entropy": 5.198822021484375,
      "epoch": 0.973103909121283,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004911254501146279,
      "loss": 5.1277,
      "mean_token_accuracy": 0.19249288439750672,
      "num_tokens": 21427315.0,
      "step": 11650
    },
    {
      "entropy": 5.277402353286743,
      "epoch": 0.973521550283996,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004911171697503193,
      "loss": 5.1766,
      "mean_token_accuracy": 0.19022970050573348,
      "num_tokens": 21437080.0,
      "step": 11655
    },
    {
      "entropy": 5.282867002487182,
      "epoch": 0.973939191446709,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004911088856026194,
      "loss": 5.0279,
      "mean_token_accuracy": 0.2144156351685524,
      "num_tokens": 21445421.0,
      "step": 11660
    },
    {
      "entropy": 5.263018560409546,
      "epoch": 0.974356832609422,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004911005976716734,
      "loss": 5.0551,
      "mean_token_accuracy": 0.20005685836076736,
      "num_tokens": 21454684.0,
      "step": 11665
    },
    {
      "entropy": 5.106145811080933,
      "epoch": 0.974774473772135,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004910923059576264,
      "loss": 5.0001,
      "mean_token_accuracy": 0.19851061701774597,
      "num_tokens": 21464287.0,
      "step": 11670
    },
    {
      "entropy": 5.211103391647339,
      "epoch": 0.975192114934848,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004910840104606234,
      "loss": 5.1423,
      "mean_token_accuracy": 0.19637335687875748,
      "num_tokens": 21472835.0,
      "step": 11675
    },
    {
      "entropy": 5.2665375709533695,
      "epoch": 0.975609756097561,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004910757111808097,
      "loss": 5.033,
      "mean_token_accuracy": 0.2001154378056526,
      "num_tokens": 21482327.0,
      "step": 11680
    },
    {
      "entropy": 5.23323621749878,
      "epoch": 0.976027397260274,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004910674081183305,
      "loss": 5.1474,
      "mean_token_accuracy": 0.18822216391563415,
      "num_tokens": 21492257.0,
      "step": 11685
    },
    {
      "entropy": 5.238407373428345,
      "epoch": 0.976445038422987,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004910591012733313,
      "loss": 5.1316,
      "mean_token_accuracy": 0.1906291127204895,
      "num_tokens": 21501930.0,
      "step": 11690
    },
    {
      "entropy": 5.347313785552979,
      "epoch": 0.9768626795857,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004910507906459575,
      "loss": 5.1394,
      "mean_token_accuracy": 0.19020719528198243,
      "num_tokens": 21511924.0,
      "step": 11695
    },
    {
      "entropy": 5.288436841964722,
      "epoch": 0.977280320748413,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004910424762363545,
      "loss": 5.1466,
      "mean_token_accuracy": 0.19041566848754882,
      "num_tokens": 21522279.0,
      "step": 11700
    },
    {
      "entropy": 5.195545864105225,
      "epoch": 0.977697961911126,
      "grad_norm": 1.3125,
      "learning_rate": 0.000491034158044668,
      "loss": 5.1217,
      "mean_token_accuracy": 0.1948204904794693,
      "num_tokens": 21532016.0,
      "step": 11705
    },
    {
      "entropy": 5.299910831451416,
      "epoch": 0.9781156030738389,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004910258360710434,
      "loss": 5.1737,
      "mean_token_accuracy": 0.18505026996135712,
      "num_tokens": 21541952.0,
      "step": 11710
    },
    {
      "entropy": 5.261684417724609,
      "epoch": 0.978533244236552,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004910175103156266,
      "loss": 4.9973,
      "mean_token_accuracy": 0.2071335181593895,
      "num_tokens": 21550185.0,
      "step": 11715
    },
    {
      "entropy": 5.262306594848633,
      "epoch": 0.978950885399265,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004910091807785632,
      "loss": 5.1622,
      "mean_token_accuracy": 0.19216437339782716,
      "num_tokens": 21558891.0,
      "step": 11720
    },
    {
      "entropy": 5.254475259780884,
      "epoch": 0.9793685265619779,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004910008474599991,
      "loss": 5.1192,
      "mean_token_accuracy": 0.1908004492521286,
      "num_tokens": 21568808.0,
      "step": 11725
    },
    {
      "entropy": 5.309691619873047,
      "epoch": 0.979786167724691,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004909925103600801,
      "loss": 5.1772,
      "mean_token_accuracy": 0.18631167560815812,
      "num_tokens": 21578838.0,
      "step": 11730
    },
    {
      "entropy": 5.205707168579101,
      "epoch": 0.980203808887404,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004909841694789523,
      "loss": 5.0429,
      "mean_token_accuracy": 0.1971341371536255,
      "num_tokens": 21588008.0,
      "step": 11735
    },
    {
      "entropy": 5.253705072402954,
      "epoch": 0.9806214500501169,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004909758248167616,
      "loss": 5.1607,
      "mean_token_accuracy": 0.191129732131958,
      "num_tokens": 21597036.0,
      "step": 11740
    },
    {
      "entropy": 5.270223140716553,
      "epoch": 0.98103909121283,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004909674763736542,
      "loss": 5.1181,
      "mean_token_accuracy": 0.19242501109838486,
      "num_tokens": 21606028.0,
      "step": 11745
    },
    {
      "entropy": 5.340626573562622,
      "epoch": 0.981456732375543,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004909591241497761,
      "loss": 5.1962,
      "mean_token_accuracy": 0.19373950511217117,
      "num_tokens": 21615377.0,
      "step": 11750
    },
    {
      "entropy": 5.349212694168091,
      "epoch": 0.9818743735382559,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004909507681452736,
      "loss": 5.2601,
      "mean_token_accuracy": 0.18453395515680313,
      "num_tokens": 21625751.0,
      "step": 11755
    },
    {
      "entropy": 5.38346037864685,
      "epoch": 0.9822920147009689,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004909424083602929,
      "loss": 5.2226,
      "mean_token_accuracy": 0.19155657887458802,
      "num_tokens": 21634825.0,
      "step": 11760
    },
    {
      "entropy": 5.298090076446533,
      "epoch": 0.982709655863682,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004909340447949805,
      "loss": 5.0689,
      "mean_token_accuracy": 0.19270638674497603,
      "num_tokens": 21644286.0,
      "step": 11765
    },
    {
      "entropy": 5.261974334716797,
      "epoch": 0.9831272970263949,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004909256774494827,
      "loss": 5.1627,
      "mean_token_accuracy": 0.19607002586126326,
      "num_tokens": 21653384.0,
      "step": 11770
    },
    {
      "entropy": 5.2690942764282225,
      "epoch": 0.9835449381891079,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000490917306323946,
      "loss": 5.0838,
      "mean_token_accuracy": 0.1957128569483757,
      "num_tokens": 21662126.0,
      "step": 11775
    },
    {
      "entropy": 5.249277925491333,
      "epoch": 0.983962579351821,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004909089314185169,
      "loss": 5.09,
      "mean_token_accuracy": 0.18836428225040436,
      "num_tokens": 21671940.0,
      "step": 11780
    },
    {
      "entropy": 5.1727166175842285,
      "epoch": 0.9843802205145339,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004909005527333421,
      "loss": 4.9988,
      "mean_token_accuracy": 0.2015775263309479,
      "num_tokens": 21680928.0,
      "step": 11785
    },
    {
      "entropy": 5.186145639419555,
      "epoch": 0.9847978616772469,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004908921702685683,
      "loss": 5.0951,
      "mean_token_accuracy": 0.19107709378004073,
      "num_tokens": 21689712.0,
      "step": 11790
    },
    {
      "entropy": 5.233446311950684,
      "epoch": 0.98521550283996,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004908837840243421,
      "loss": 5.0589,
      "mean_token_accuracy": 0.19764397740364076,
      "num_tokens": 21698325.0,
      "step": 11795
    },
    {
      "entropy": 5.193361043930054,
      "epoch": 0.9856331440026729,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004908753940008105,
      "loss": 5.0341,
      "mean_token_accuracy": 0.20194266140460967,
      "num_tokens": 21707607.0,
      "step": 11800
    },
    {
      "entropy": 5.136324787139893,
      "epoch": 0.9860507851653859,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004908670001981201,
      "loss": 5.0088,
      "mean_token_accuracy": 0.20002978593111037,
      "num_tokens": 21715887.0,
      "step": 11805
    },
    {
      "entropy": 5.231118202209473,
      "epoch": 0.986468426328099,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004908586026164182,
      "loss": 5.1377,
      "mean_token_accuracy": 0.19563366025686263,
      "num_tokens": 21724992.0,
      "step": 11810
    },
    {
      "entropy": 5.280947256088257,
      "epoch": 0.9868860674908119,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004908502012558516,
      "loss": 5.1251,
      "mean_token_accuracy": 0.1963654264807701,
      "num_tokens": 21734317.0,
      "step": 11815
    },
    {
      "entropy": 5.2773841381072994,
      "epoch": 0.9873037086535249,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004908417961165674,
      "loss": 5.1729,
      "mean_token_accuracy": 0.19197501242160797,
      "num_tokens": 21743014.0,
      "step": 11820
    },
    {
      "entropy": 5.31586799621582,
      "epoch": 0.9877213498162379,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004908333871987126,
      "loss": 5.0875,
      "mean_token_accuracy": 0.19767518043518068,
      "num_tokens": 21751058.0,
      "step": 11825
    },
    {
      "entropy": 5.294144153594971,
      "epoch": 0.9881389909789509,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004908249745024347,
      "loss": 5.1867,
      "mean_token_accuracy": 0.19202081561088563,
      "num_tokens": 21760549.0,
      "step": 11830
    },
    {
      "entropy": 5.245868825912476,
      "epoch": 0.9885566321416639,
      "grad_norm": 1.375,
      "learning_rate": 0.000490816558027881,
      "loss": 5.1021,
      "mean_token_accuracy": 0.20167008489370347,
      "num_tokens": 21769873.0,
      "step": 11835
    },
    {
      "entropy": 5.2595337390899655,
      "epoch": 0.9889742733043769,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004908081377751985,
      "loss": 5.1944,
      "mean_token_accuracy": 0.19685014337301254,
      "num_tokens": 21779036.0,
      "step": 11840
    },
    {
      "entropy": 5.395480632781982,
      "epoch": 0.9893919144670899,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004907997137445347,
      "loss": 5.1876,
      "mean_token_accuracy": 0.18601257354021072,
      "num_tokens": 21787671.0,
      "step": 11845
    },
    {
      "entropy": 5.329554176330566,
      "epoch": 0.9898095556298029,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004907912859360373,
      "loss": 5.139,
      "mean_token_accuracy": 0.1899839907884598,
      "num_tokens": 21797269.0,
      "step": 11850
    },
    {
      "entropy": 5.310826587677002,
      "epoch": 0.9902271967925159,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004907828543498536,
      "loss": 5.2024,
      "mean_token_accuracy": 0.19194122701883315,
      "num_tokens": 21806738.0,
      "step": 11855
    },
    {
      "entropy": 5.3051004886627195,
      "epoch": 0.9906448379552288,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004907744189861313,
      "loss": 5.1041,
      "mean_token_accuracy": 0.19699872732162477,
      "num_tokens": 21815612.0,
      "step": 11860
    },
    {
      "entropy": 5.207951354980469,
      "epoch": 0.9910624791179419,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000490765979845018,
      "loss": 5.1247,
      "mean_token_accuracy": 0.2008882626891136,
      "num_tokens": 21824222.0,
      "step": 11865
    },
    {
      "entropy": 5.326421308517456,
      "epoch": 0.9914801202806549,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004907575369266617,
      "loss": 5.1299,
      "mean_token_accuracy": 0.19537571519613267,
      "num_tokens": 21833929.0,
      "step": 11870
    },
    {
      "entropy": 5.322073888778687,
      "epoch": 0.9918977614433678,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004907490902312099,
      "loss": 5.1661,
      "mean_token_accuracy": 0.19690808057785034,
      "num_tokens": 21843261.0,
      "step": 11875
    },
    {
      "entropy": 5.321474361419678,
      "epoch": 0.9923154026060809,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004907406397588106,
      "loss": 5.2169,
      "mean_token_accuracy": 0.1859991580247879,
      "num_tokens": 21853004.0,
      "step": 11880
    },
    {
      "entropy": 5.2847131252288815,
      "epoch": 0.9927330437687939,
      "grad_norm": 1.375,
      "learning_rate": 0.0004907321855096117,
      "loss": 5.0617,
      "mean_token_accuracy": 0.19275859892368316,
      "num_tokens": 21862628.0,
      "step": 11885
    },
    {
      "entropy": 5.199124813079834,
      "epoch": 0.9931506849315068,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004907237274837613,
      "loss": 5.0529,
      "mean_token_accuracy": 0.1993335708975792,
      "num_tokens": 21872041.0,
      "step": 11890
    },
    {
      "entropy": 5.239595603942871,
      "epoch": 0.9935683260942199,
      "grad_norm": 1.25,
      "learning_rate": 0.0004907152656814072,
      "loss": 5.148,
      "mean_token_accuracy": 0.18845521211624144,
      "num_tokens": 21881834.0,
      "step": 11895
    },
    {
      "entropy": 5.371371507644653,
      "epoch": 0.9939859672569329,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004907068001026978,
      "loss": 5.2403,
      "mean_token_accuracy": 0.1892232820391655,
      "num_tokens": 21892349.0,
      "step": 11900
    },
    {
      "entropy": 5.30948224067688,
      "epoch": 0.9944036084196458,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004906983307477813,
      "loss": 5.1188,
      "mean_token_accuracy": 0.1962111160159111,
      "num_tokens": 21901427.0,
      "step": 11905
    },
    {
      "entropy": 5.279951047897339,
      "epoch": 0.9948212495823588,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004906898576168057,
      "loss": 5.0619,
      "mean_token_accuracy": 0.19501851201057435,
      "num_tokens": 21910654.0,
      "step": 11910
    },
    {
      "entropy": 5.132517528533936,
      "epoch": 0.9952388907450719,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004906813807099197,
      "loss": 5.0668,
      "mean_token_accuracy": 0.19910750091075896,
      "num_tokens": 21919583.0,
      "step": 11915
    },
    {
      "entropy": 5.162035608291626,
      "epoch": 0.9956565319077848,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004906729000272714,
      "loss": 4.9971,
      "mean_token_accuracy": 0.2020763099193573,
      "num_tokens": 21927843.0,
      "step": 11920
    },
    {
      "entropy": 5.252127265930175,
      "epoch": 0.9960741730704978,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004906644155690095,
      "loss": 5.0619,
      "mean_token_accuracy": 0.1962496042251587,
      "num_tokens": 21937269.0,
      "step": 11925
    },
    {
      "entropy": 5.4172626495361325,
      "epoch": 0.9964918142332109,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004906559273352823,
      "loss": 5.2525,
      "mean_token_accuracy": 0.18500997871160507,
      "num_tokens": 21946965.0,
      "step": 11930
    },
    {
      "entropy": 5.258734655380249,
      "epoch": 0.9969094553959238,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004906474353262386,
      "loss": 5.0801,
      "mean_token_accuracy": 0.19003722816705704,
      "num_tokens": 21955957.0,
      "step": 11935
    },
    {
      "entropy": 5.190324974060059,
      "epoch": 0.9973270965586368,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004906389395420269,
      "loss": 5.0271,
      "mean_token_accuracy": 0.2002367466688156,
      "num_tokens": 21965643.0,
      "step": 11940
    },
    {
      "entropy": 5.24493088722229,
      "epoch": 0.9977447377213499,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000490630439982796,
      "loss": 5.0972,
      "mean_token_accuracy": 0.19855068027973174,
      "num_tokens": 21974789.0,
      "step": 11945
    },
    {
      "entropy": 5.194629096984864,
      "epoch": 0.9981623788840628,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004906219366486947,
      "loss": 5.0083,
      "mean_token_accuracy": 0.20713224560022353,
      "num_tokens": 21983411.0,
      "step": 11950
    },
    {
      "entropy": 5.236247491836548,
      "epoch": 0.9985800200467758,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004906134295398719,
      "loss": 5.1229,
      "mean_token_accuracy": 0.1941255658864975,
      "num_tokens": 21992570.0,
      "step": 11955
    },
    {
      "entropy": 5.134074306488037,
      "epoch": 0.9989976612094889,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004906049186564764,
      "loss": 4.9635,
      "mean_token_accuracy": 0.20698622912168502,
      "num_tokens": 22001411.0,
      "step": 11960
    },
    {
      "entropy": 5.37010383605957,
      "epoch": 0.9994153023722018,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004905964039986574,
      "loss": 5.2244,
      "mean_token_accuracy": 0.18307197839021683,
      "num_tokens": 22010882.0,
      "step": 11965
    },
    {
      "entropy": 5.305927276611328,
      "epoch": 0.9998329435349148,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004905878855665637,
      "loss": 5.2001,
      "mean_token_accuracy": 0.1850425347685814,
      "num_tokens": 22020885.0,
      "step": 11970
    },
    {
      "entropy": 5.3142174243927,
      "epoch": 1.0002505846976277,
      "grad_norm": 1.25,
      "learning_rate": 0.0004905793633603446,
      "loss": 4.9662,
      "mean_token_accuracy": 0.20107216835021974,
      "num_tokens": 22030278.0,
      "step": 11975
    },
    {
      "entropy": 5.2956095218658445,
      "epoch": 1.0006682258603408,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004905708373801492,
      "loss": 5.1518,
      "mean_token_accuracy": 0.18832548558712006,
      "num_tokens": 22039601.0,
      "step": 11980
    },
    {
      "entropy": 5.139204502105713,
      "epoch": 1.0010858670230538,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004905623076261269,
      "loss": 4.9953,
      "mean_token_accuracy": 0.20047934651374816,
      "num_tokens": 22049158.0,
      "step": 11985
    },
    {
      "entropy": 5.276647758483887,
      "epoch": 1.0015035081857668,
      "grad_norm": 1.3125,
      "learning_rate": 0.000490553774098427,
      "loss": 5.1011,
      "mean_token_accuracy": 0.18474566340446472,
      "num_tokens": 22059278.0,
      "step": 11990
    },
    {
      "entropy": 5.306740760803223,
      "epoch": 1.0019211493484799,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004905452367971986,
      "loss": 5.0551,
      "mean_token_accuracy": 0.19967539757490158,
      "num_tokens": 22068415.0,
      "step": 11995
    },
    {
      "entropy": 5.271653747558593,
      "epoch": 1.0023387905111927,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004905366957225916,
      "loss": 5.0276,
      "mean_token_accuracy": 0.19304035753011703,
      "num_tokens": 22077815.0,
      "step": 12000
    },
    {
      "epoch": 1.0023387905111927,
      "eval_entropy": 5.005528071148369,
      "eval_loss": 5.074009895324707,
      "eval_mean_token_accuracy": 0.20515256969258583,
      "eval_num_tokens": 22077815.0,
      "eval_runtime": 27.6569,
      "eval_samples_per_second": 1365.412,
      "eval_steps_per_second": 170.699,
      "step": 12000
    },
    {
      "entropy": 5.272951555252075,
      "epoch": 1.0027564316739057,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004905281508747552,
      "loss": 5.1021,
      "mean_token_accuracy": 0.18844978958368303,
      "num_tokens": 22087606.0,
      "step": 12005
    },
    {
      "entropy": 5.193337440490723,
      "epoch": 1.0031740728366187,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004905196022538391,
      "loss": 4.9464,
      "mean_token_accuracy": 0.20142873078584672,
      "num_tokens": 22096300.0,
      "step": 12010
    },
    {
      "entropy": 5.295071220397949,
      "epoch": 1.0035917139993318,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000490511049859993,
      "loss": 5.1297,
      "mean_token_accuracy": 0.19523701071739197,
      "num_tokens": 22105750.0,
      "step": 12015
    },
    {
      "entropy": 5.316681957244873,
      "epoch": 1.0040093551620448,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004905024936933664,
      "loss": 5.1209,
      "mean_token_accuracy": 0.1969589725136757,
      "num_tokens": 22115163.0,
      "step": 12020
    },
    {
      "entropy": 5.2998124122619625,
      "epoch": 1.0044269963247578,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004904939337541093,
      "loss": 5.1109,
      "mean_token_accuracy": 0.19299143105745314,
      "num_tokens": 22125580.0,
      "step": 12025
    },
    {
      "entropy": 5.156743001937866,
      "epoch": 1.0048446374874707,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004904853700423716,
      "loss": 4.9858,
      "mean_token_accuracy": 0.20624399185180664,
      "num_tokens": 22134985.0,
      "step": 12030
    },
    {
      "entropy": 5.295426082611084,
      "epoch": 1.0052622786501837,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004904768025583031,
      "loss": 4.9776,
      "mean_token_accuracy": 0.20252305120229722,
      "num_tokens": 22142530.0,
      "step": 12035
    },
    {
      "entropy": 5.285527181625366,
      "epoch": 1.0056799198128967,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004904682313020537,
      "loss": 5.0891,
      "mean_token_accuracy": 0.19556759148836136,
      "num_tokens": 22151696.0,
      "step": 12040
    },
    {
      "entropy": 5.184085369110107,
      "epoch": 1.0060975609756098,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004904596562737737,
      "loss": 5.0254,
      "mean_token_accuracy": 0.195164991915226,
      "num_tokens": 22161286.0,
      "step": 12045
    },
    {
      "entropy": 5.207891178131104,
      "epoch": 1.0065152021383228,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000490451077473613,
      "loss": 4.9902,
      "mean_token_accuracy": 0.2023245081305504,
      "num_tokens": 22169678.0,
      "step": 12050
    },
    {
      "entropy": 5.225547218322754,
      "epoch": 1.0069328433010358,
      "grad_norm": 1.375,
      "learning_rate": 0.0004904424949017218,
      "loss": 5.0378,
      "mean_token_accuracy": 0.20860767960548401,
      "num_tokens": 22178618.0,
      "step": 12055
    },
    {
      "entropy": 5.185265731811524,
      "epoch": 1.0073504844637486,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004904339085582504,
      "loss": 5.0062,
      "mean_token_accuracy": 0.2004217177629471,
      "num_tokens": 22187574.0,
      "step": 12060
    },
    {
      "entropy": 5.278010320663452,
      "epoch": 1.0077681256264617,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004904253184433491,
      "loss": 5.1795,
      "mean_token_accuracy": 0.18732508569955825,
      "num_tokens": 22198840.0,
      "step": 12065
    },
    {
      "entropy": 5.220087337493896,
      "epoch": 1.0081857667891747,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004904167245571684,
      "loss": 4.9775,
      "mean_token_accuracy": 0.20756881833076476,
      "num_tokens": 22207166.0,
      "step": 12070
    },
    {
      "entropy": 5.223711919784546,
      "epoch": 1.0086034079518877,
      "grad_norm": 1.25,
      "learning_rate": 0.0004904081268998586,
      "loss": 5.0034,
      "mean_token_accuracy": 0.18989900052547454,
      "num_tokens": 22216232.0,
      "step": 12075
    },
    {
      "entropy": 5.209487390518189,
      "epoch": 1.0090210491146008,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004903995254715702,
      "loss": 4.9847,
      "mean_token_accuracy": 0.19870017766952514,
      "num_tokens": 22224853.0,
      "step": 12080
    },
    {
      "entropy": 5.152215003967285,
      "epoch": 1.0094386902773138,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004903909202724539,
      "loss": 4.9734,
      "mean_token_accuracy": 0.20457060039043426,
      "num_tokens": 22233971.0,
      "step": 12085
    },
    {
      "entropy": 5.149052238464355,
      "epoch": 1.0098563314400266,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004903823113026602,
      "loss": 4.8796,
      "mean_token_accuracy": 0.20572237223386763,
      "num_tokens": 22242548.0,
      "step": 12090
    },
    {
      "entropy": 5.073572206497192,
      "epoch": 1.0102739726027397,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00049037369856234,
      "loss": 4.854,
      "mean_token_accuracy": 0.21345870047807694,
      "num_tokens": 22251086.0,
      "step": 12095
    },
    {
      "entropy": 5.221481084823608,
      "epoch": 1.0106916137654527,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004903650820516438,
      "loss": 5.0499,
      "mean_token_accuracy": 0.19278639554977417,
      "num_tokens": 22260035.0,
      "step": 12100
    },
    {
      "entropy": 5.230033731460571,
      "epoch": 1.0111092549281657,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004903564617707227,
      "loss": 4.9841,
      "mean_token_accuracy": 0.20316994041204453,
      "num_tokens": 22269134.0,
      "step": 12105
    },
    {
      "entropy": 5.258654260635376,
      "epoch": 1.0115268960908788,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004903478377197275,
      "loss": 5.0638,
      "mean_token_accuracy": 0.2021499365568161,
      "num_tokens": 22278123.0,
      "step": 12110
    },
    {
      "entropy": 5.199178838729859,
      "epoch": 1.0119445372535918,
      "grad_norm": 1.375,
      "learning_rate": 0.0004903392098988092,
      "loss": 5.0741,
      "mean_token_accuracy": 0.1921963483095169,
      "num_tokens": 22287557.0,
      "step": 12115
    },
    {
      "entropy": 5.223395204544067,
      "epoch": 1.0123621784163046,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004903305783081188,
      "loss": 5.0134,
      "mean_token_accuracy": 0.19927221089601516,
      "num_tokens": 22296705.0,
      "step": 12120
    },
    {
      "entropy": 5.1888343334198,
      "epoch": 1.0127798195790176,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004903219429478075,
      "loss": 4.9716,
      "mean_token_accuracy": 0.1977076917886734,
      "num_tokens": 22305904.0,
      "step": 12125
    },
    {
      "entropy": 5.208519983291626,
      "epoch": 1.0131974607417307,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004903133038180263,
      "loss": 5.0123,
      "mean_token_accuracy": 0.1983116537332535,
      "num_tokens": 22315051.0,
      "step": 12130
    },
    {
      "entropy": 5.219885015487671,
      "epoch": 1.0136151019044437,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004903046609189266,
      "loss": 5.0615,
      "mean_token_accuracy": 0.1974531054496765,
      "num_tokens": 22324469.0,
      "step": 12135
    },
    {
      "entropy": 5.36426191329956,
      "epoch": 1.0140327430671567,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004902960142506597,
      "loss": 5.1316,
      "mean_token_accuracy": 0.19027840346097946,
      "num_tokens": 22334407.0,
      "step": 12140
    },
    {
      "entropy": 5.305349588394165,
      "epoch": 1.0144503842298698,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004902873638133768,
      "loss": 5.1301,
      "mean_token_accuracy": 0.18950580656528473,
      "num_tokens": 22342975.0,
      "step": 12145
    },
    {
      "entropy": 5.197851753234863,
      "epoch": 1.0148680253925826,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004902787096072296,
      "loss": 4.9476,
      "mean_token_accuracy": 0.19511666297912597,
      "num_tokens": 22351649.0,
      "step": 12150
    },
    {
      "entropy": 5.228479242324829,
      "epoch": 1.0152856665552956,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004902700516323693,
      "loss": 5.0444,
      "mean_token_accuracy": 0.18938060253858566,
      "num_tokens": 22361956.0,
      "step": 12155
    },
    {
      "entropy": 5.237329149246216,
      "epoch": 1.0157033077180087,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004902613898889477,
      "loss": 5.0382,
      "mean_token_accuracy": 0.19228207170963288,
      "num_tokens": 22372354.0,
      "step": 12160
    },
    {
      "entropy": 5.168973588943482,
      "epoch": 1.0161209488807217,
      "grad_norm": 1.25,
      "learning_rate": 0.0004902527243771164,
      "loss": 4.9104,
      "mean_token_accuracy": 0.1978227287530899,
      "num_tokens": 22381225.0,
      "step": 12165
    },
    {
      "entropy": 5.09970874786377,
      "epoch": 1.0165385900434347,
      "grad_norm": 1.328125,
      "learning_rate": 0.000490244055097027,
      "loss": 4.9383,
      "mean_token_accuracy": 0.20618234276771547,
      "num_tokens": 22390666.0,
      "step": 12170
    },
    {
      "entropy": 5.148920106887817,
      "epoch": 1.0169562312061478,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004902353820488314,
      "loss": 4.9364,
      "mean_token_accuracy": 0.20314700603485109,
      "num_tokens": 22399528.0,
      "step": 12175
    },
    {
      "entropy": 5.1705070495605465,
      "epoch": 1.0173738723688606,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004902267052326812,
      "loss": 4.9613,
      "mean_token_accuracy": 0.20630052089691162,
      "num_tokens": 22408827.0,
      "step": 12180
    },
    {
      "entropy": 5.175075054168701,
      "epoch": 1.0177915135315736,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004902180246487285,
      "loss": 5.0075,
      "mean_token_accuracy": 0.19284019023180007,
      "num_tokens": 22418098.0,
      "step": 12185
    },
    {
      "entropy": 5.104143905639648,
      "epoch": 1.0182091546942866,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004902093402971254,
      "loss": 4.864,
      "mean_token_accuracy": 0.20585375130176545,
      "num_tokens": 22427417.0,
      "step": 12190
    },
    {
      "entropy": 5.191632890701294,
      "epoch": 1.0186267958569997,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004902006521780236,
      "loss": 5.0507,
      "mean_token_accuracy": 0.19742044657468796,
      "num_tokens": 22436938.0,
      "step": 12195
    },
    {
      "entropy": 5.209488153457642,
      "epoch": 1.0190444370197127,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004901919602915754,
      "loss": 5.0911,
      "mean_token_accuracy": 0.1913207560777664,
      "num_tokens": 22447361.0,
      "step": 12200
    },
    {
      "entropy": 5.313965654373169,
      "epoch": 1.0194620781824257,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004901832646379328,
      "loss": 5.1051,
      "mean_token_accuracy": 0.19502849578857423,
      "num_tokens": 22457431.0,
      "step": 12205
    },
    {
      "entropy": 5.258908224105835,
      "epoch": 1.0198797193451385,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004901745652172483,
      "loss": 5.0921,
      "mean_token_accuracy": 0.197348390519619,
      "num_tokens": 22466219.0,
      "step": 12210
    },
    {
      "entropy": 5.311530303955078,
      "epoch": 1.0202973605078516,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000490165862029674,
      "loss": 5.1222,
      "mean_token_accuracy": 0.19293407946825028,
      "num_tokens": 22476231.0,
      "step": 12215
    },
    {
      "entropy": 5.174165868759156,
      "epoch": 1.0207150016705646,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004901571550753622,
      "loss": 4.899,
      "mean_token_accuracy": 0.20714213848114013,
      "num_tokens": 22484288.0,
      "step": 12220
    },
    {
      "entropy": 5.185079097747803,
      "epoch": 1.0211326428332776,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004901484443544655,
      "loss": 5.0516,
      "mean_token_accuracy": 0.20046135187149047,
      "num_tokens": 22493619.0,
      "step": 12225
    },
    {
      "entropy": 5.102109289169311,
      "epoch": 1.0215502839959907,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004901397298671362,
      "loss": 4.949,
      "mean_token_accuracy": 0.20625572204589843,
      "num_tokens": 22501606.0,
      "step": 12230
    },
    {
      "entropy": 5.197969675064087,
      "epoch": 1.0219679251587037,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004901310116135271,
      "loss": 4.9557,
      "mean_token_accuracy": 0.19927047938108444,
      "num_tokens": 22510410.0,
      "step": 12235
    },
    {
      "entropy": 5.194157123565674,
      "epoch": 1.0223855663214167,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004901222895937905,
      "loss": 5.0752,
      "mean_token_accuracy": 0.2020103618502617,
      "num_tokens": 22518796.0,
      "step": 12240
    },
    {
      "entropy": 5.175445222854615,
      "epoch": 1.0228032074841296,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004901135638080796,
      "loss": 4.8887,
      "mean_token_accuracy": 0.20532342344522475,
      "num_tokens": 22527532.0,
      "step": 12245
    },
    {
      "entropy": 5.142603063583374,
      "epoch": 1.0232208486468426,
      "grad_norm": 1.375,
      "learning_rate": 0.0004901048342565465,
      "loss": 4.979,
      "mean_token_accuracy": 0.19879175573587418,
      "num_tokens": 22536882.0,
      "step": 12250
    },
    {
      "entropy": 5.16430630683899,
      "epoch": 1.0236384898095556,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004900961009393445,
      "loss": 4.8902,
      "mean_token_accuracy": 0.20133597552776336,
      "num_tokens": 22545549.0,
      "step": 12255
    },
    {
      "entropy": 5.245450162887574,
      "epoch": 1.0240561309722687,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004900873638566263,
      "loss": 5.1105,
      "mean_token_accuracy": 0.1916583687067032,
      "num_tokens": 22554941.0,
      "step": 12260
    },
    {
      "entropy": 5.180256271362305,
      "epoch": 1.0244737721349817,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000490078623008545,
      "loss": 5.0215,
      "mean_token_accuracy": 0.196419857442379,
      "num_tokens": 22565159.0,
      "step": 12265
    },
    {
      "entropy": 5.129810428619384,
      "epoch": 1.0248914132976945,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004900698783952533,
      "loss": 4.9535,
      "mean_token_accuracy": 0.20413996577262877,
      "num_tokens": 22574184.0,
      "step": 12270
    },
    {
      "entropy": 5.201740026473999,
      "epoch": 1.0253090544604075,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004900611300169048,
      "loss": 5.0515,
      "mean_token_accuracy": 0.19606004208326339,
      "num_tokens": 22584062.0,
      "step": 12275
    },
    {
      "entropy": 5.235516786575317,
      "epoch": 1.0257266956231206,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004900523778736521,
      "loss": 4.9862,
      "mean_token_accuracy": 0.19683620929718018,
      "num_tokens": 22592845.0,
      "step": 12280
    },
    {
      "entropy": 5.210376882553101,
      "epoch": 1.0261443367858336,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004900436219656488,
      "loss": 5.0259,
      "mean_token_accuracy": 0.19351164996623993,
      "num_tokens": 22602750.0,
      "step": 12285
    },
    {
      "entropy": 5.188666486740113,
      "epoch": 1.0265619779485466,
      "grad_norm": 1.234375,
      "learning_rate": 0.000490034862293048,
      "loss": 5.0374,
      "mean_token_accuracy": 0.19255138486623763,
      "num_tokens": 22611788.0,
      "step": 12290
    },
    {
      "entropy": 5.167825174331665,
      "epoch": 1.0269796191112597,
      "grad_norm": 1.5,
      "learning_rate": 0.0004900260988560031,
      "loss": 4.9551,
      "mean_token_accuracy": 0.1969916671514511,
      "num_tokens": 22620234.0,
      "step": 12295
    },
    {
      "entropy": 5.119522905349731,
      "epoch": 1.0273972602739727,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004900173316546675,
      "loss": 4.9007,
      "mean_token_accuracy": 0.20260113775730132,
      "num_tokens": 22629155.0,
      "step": 12300
    },
    {
      "entropy": 5.2697563648223875,
      "epoch": 1.0278149014366855,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004900085606891947,
      "loss": 5.1127,
      "mean_token_accuracy": 0.18663983196020126,
      "num_tokens": 22638647.0,
      "step": 12305
    },
    {
      "entropy": 5.264565515518188,
      "epoch": 1.0282325425993986,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004899997859597382,
      "loss": 5.0113,
      "mean_token_accuracy": 0.19255275875329972,
      "num_tokens": 22648295.0,
      "step": 12310
    },
    {
      "entropy": 5.170897054672241,
      "epoch": 1.0286501837621116,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004899910074664518,
      "loss": 5.0174,
      "mean_token_accuracy": 0.1934913292527199,
      "num_tokens": 22657342.0,
      "step": 12315
    },
    {
      "entropy": 5.163343286514282,
      "epoch": 1.0290678249248246,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004899822252094889,
      "loss": 4.968,
      "mean_token_accuracy": 0.20334915071725845,
      "num_tokens": 22665230.0,
      "step": 12320
    },
    {
      "entropy": 5.156237936019897,
      "epoch": 1.0294854660875377,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004899734391890034,
      "loss": 4.9989,
      "mean_token_accuracy": 0.19859781861305237,
      "num_tokens": 22674385.0,
      "step": 12325
    },
    {
      "entropy": 5.112623071670532,
      "epoch": 1.0299031072502507,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004899646494051491,
      "loss": 4.8647,
      "mean_token_accuracy": 0.21570334136486052,
      "num_tokens": 22683493.0,
      "step": 12330
    },
    {
      "entropy": 5.146523904800415,
      "epoch": 1.0303207484129635,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004899558558580798,
      "loss": 4.9372,
      "mean_token_accuracy": 0.19660401940345765,
      "num_tokens": 22691878.0,
      "step": 12335
    },
    {
      "entropy": 5.093887996673584,
      "epoch": 1.0307383895756765,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004899470585479495,
      "loss": 4.948,
      "mean_token_accuracy": 0.1973460853099823,
      "num_tokens": 22700112.0,
      "step": 12340
    },
    {
      "entropy": 5.172760009765625,
      "epoch": 1.0311560307383896,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004899382574749124,
      "loss": 4.984,
      "mean_token_accuracy": 0.1959293529391289,
      "num_tokens": 22709218.0,
      "step": 12345
    },
    {
      "entropy": 5.21093716621399,
      "epoch": 1.0315736719011026,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004899294526391221,
      "loss": 4.9983,
      "mean_token_accuracy": 0.2043227106332779,
      "num_tokens": 22718423.0,
      "step": 12350
    },
    {
      "entropy": 5.155216407775879,
      "epoch": 1.0319913130638156,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004899206440407332,
      "loss": 5.0827,
      "mean_token_accuracy": 0.1896908238530159,
      "num_tokens": 22727333.0,
      "step": 12355
    },
    {
      "entropy": 5.226324558258057,
      "epoch": 1.0324089542265287,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004899118316798996,
      "loss": 5.0043,
      "mean_token_accuracy": 0.19321927577257156,
      "num_tokens": 22737248.0,
      "step": 12360
    },
    {
      "entropy": 5.186983489990235,
      "epoch": 1.0328265953892415,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004899030155567759,
      "loss": 5.0208,
      "mean_token_accuracy": 0.1934382826089859,
      "num_tokens": 22747515.0,
      "step": 12365
    },
    {
      "entropy": 5.264181184768677,
      "epoch": 1.0332442365519545,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000489894195671516,
      "loss": 5.0145,
      "mean_token_accuracy": 0.20166782140731812,
      "num_tokens": 22756670.0,
      "step": 12370
    },
    {
      "entropy": 5.228320741653443,
      "epoch": 1.0336618777146676,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004898853720242747,
      "loss": 5.1555,
      "mean_token_accuracy": 0.184278404712677,
      "num_tokens": 22766332.0,
      "step": 12375
    },
    {
      "entropy": 5.230732488632202,
      "epoch": 1.0340795188773806,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004898765446152062,
      "loss": 4.9836,
      "mean_token_accuracy": 0.20058978945016862,
      "num_tokens": 22774999.0,
      "step": 12380
    },
    {
      "entropy": 5.169930696487427,
      "epoch": 1.0344971600400936,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004898677134444652,
      "loss": 5.0079,
      "mean_token_accuracy": 0.20046788901090623,
      "num_tokens": 22784578.0,
      "step": 12385
    },
    {
      "entropy": 5.18506965637207,
      "epoch": 1.0349148012028067,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004898588785122063,
      "loss": 5.0053,
      "mean_token_accuracy": 0.1997951254248619,
      "num_tokens": 22794508.0,
      "step": 12390
    },
    {
      "entropy": 5.247910690307617,
      "epoch": 1.0353324423655195,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004898500398185841,
      "loss": 5.1318,
      "mean_token_accuracy": 0.19153245240449907,
      "num_tokens": 22804399.0,
      "step": 12395
    },
    {
      "entropy": 5.24885663986206,
      "epoch": 1.0357500835282325,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004898411973637532,
      "loss": 5.0206,
      "mean_token_accuracy": 0.1955435261130333,
      "num_tokens": 22814361.0,
      "step": 12400
    },
    {
      "entropy": 5.141165113449096,
      "epoch": 1.0361677246909455,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004898323511478686,
      "loss": 4.9562,
      "mean_token_accuracy": 0.2015901431441307,
      "num_tokens": 22823780.0,
      "step": 12405
    },
    {
      "entropy": 5.171428108215332,
      "epoch": 1.0365853658536586,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004898235011710851,
      "loss": 4.9553,
      "mean_token_accuracy": 0.19667875319719313,
      "num_tokens": 22832575.0,
      "step": 12410
    },
    {
      "entropy": 5.076723718643189,
      "epoch": 1.0370030070163716,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004898146474335577,
      "loss": 4.8112,
      "mean_token_accuracy": 0.21225759387016296,
      "num_tokens": 22841822.0,
      "step": 12415
    },
    {
      "entropy": 5.296346187591553,
      "epoch": 1.0374206481790846,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004898057899354412,
      "loss": 5.0879,
      "mean_token_accuracy": 0.19025665074586867,
      "num_tokens": 22851835.0,
      "step": 12420
    },
    {
      "entropy": 5.160854053497315,
      "epoch": 1.0378382893417974,
      "grad_norm": 1.234375,
      "learning_rate": 0.000489796928676891,
      "loss": 5.0106,
      "mean_token_accuracy": 0.19954221397638322,
      "num_tokens": 22861025.0,
      "step": 12425
    },
    {
      "entropy": 5.182194471359253,
      "epoch": 1.0382559305045105,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004897880636580618,
      "loss": 5.0109,
      "mean_token_accuracy": 0.19988245964050294,
      "num_tokens": 22869721.0,
      "step": 12430
    },
    {
      "entropy": 5.165585708618164,
      "epoch": 1.0386735716672235,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000489779194879109,
      "loss": 4.9014,
      "mean_token_accuracy": 0.21120780408382417,
      "num_tokens": 22878225.0,
      "step": 12435
    },
    {
      "entropy": 5.207777833938598,
      "epoch": 1.0390912128299366,
      "grad_norm": 1.21875,
      "learning_rate": 0.000489770322340188,
      "loss": 5.1331,
      "mean_token_accuracy": 0.19234661906957626,
      "num_tokens": 22888598.0,
      "step": 12440
    },
    {
      "entropy": 5.247703552246094,
      "epoch": 1.0395088539926496,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004897614460414539,
      "loss": 5.0379,
      "mean_token_accuracy": 0.19612369388341905,
      "num_tokens": 22897326.0,
      "step": 12445
    },
    {
      "entropy": 5.20391993522644,
      "epoch": 1.0399264951553626,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004897525659830623,
      "loss": 5.0267,
      "mean_token_accuracy": 0.19482456594705583,
      "num_tokens": 22906372.0,
      "step": 12450
    },
    {
      "entropy": 5.1094457626342775,
      "epoch": 1.0403441363180754,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004897436821651683,
      "loss": 4.9131,
      "mean_token_accuracy": 0.20987346172332763,
      "num_tokens": 22915987.0,
      "step": 12455
    },
    {
      "entropy": 5.22169246673584,
      "epoch": 1.0407617774807885,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004897347945879278,
      "loss": 5.0505,
      "mean_token_accuracy": 0.19595839828252792,
      "num_tokens": 22926118.0,
      "step": 12460
    },
    {
      "entropy": 5.19300012588501,
      "epoch": 1.0411794186435015,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004897259032514963,
      "loss": 4.9904,
      "mean_token_accuracy": 0.1996393769979477,
      "num_tokens": 22934813.0,
      "step": 12465
    },
    {
      "entropy": 5.199761199951172,
      "epoch": 1.0415970598062145,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004897170081560292,
      "loss": 5.1025,
      "mean_token_accuracy": 0.19380462616682054,
      "num_tokens": 22944579.0,
      "step": 12470
    },
    {
      "entropy": 5.131135416030884,
      "epoch": 1.0420147009689276,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004897081093016827,
      "loss": 4.9239,
      "mean_token_accuracy": 0.20497809499502181,
      "num_tokens": 22953515.0,
      "step": 12475
    },
    {
      "entropy": 5.204380798339844,
      "epoch": 1.0424323421316406,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004896992066886121,
      "loss": 5.068,
      "mean_token_accuracy": 0.19150227457284927,
      "num_tokens": 22962599.0,
      "step": 12480
    },
    {
      "entropy": 5.225791788101196,
      "epoch": 1.0428499832943534,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004896903003169735,
      "loss": 5.0438,
      "mean_token_accuracy": 0.20676450729370116,
      "num_tokens": 22971717.0,
      "step": 12485
    },
    {
      "entropy": 5.092175912857056,
      "epoch": 1.0432676244570664,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004896813901869229,
      "loss": 4.8157,
      "mean_token_accuracy": 0.20538005530834197,
      "num_tokens": 22980587.0,
      "step": 12490
    },
    {
      "entropy": 5.179116487503052,
      "epoch": 1.0436852656197795,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004896724762986159,
      "loss": 5.0089,
      "mean_token_accuracy": 0.20922803431749343,
      "num_tokens": 22990197.0,
      "step": 12495
    },
    {
      "entropy": 5.214642190933228,
      "epoch": 1.0441029067824925,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004896635586522089,
      "loss": 5.0649,
      "mean_token_accuracy": 0.19622240960597992,
      "num_tokens": 22999670.0,
      "step": 12500
    },
    {
      "entropy": 5.238640356063843,
      "epoch": 1.0445205479452055,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004896546372478581,
      "loss": 5.0134,
      "mean_token_accuracy": 0.20502431243658065,
      "num_tokens": 23008125.0,
      "step": 12505
    },
    {
      "entropy": 5.13493537902832,
      "epoch": 1.0449381891079186,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004896457120857192,
      "loss": 4.8861,
      "mean_token_accuracy": 0.20898921489715577,
      "num_tokens": 23017469.0,
      "step": 12510
    },
    {
      "entropy": 5.1521985054016115,
      "epoch": 1.0453558302706314,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000489636783165949,
      "loss": 5.0224,
      "mean_token_accuracy": 0.19859592169523238,
      "num_tokens": 23026468.0,
      "step": 12515
    },
    {
      "entropy": 5.283107757568359,
      "epoch": 1.0457734714333444,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004896278504887033,
      "loss": 5.0631,
      "mean_token_accuracy": 0.20096652060747147,
      "num_tokens": 23034944.0,
      "step": 12520
    },
    {
      "entropy": 5.135332107543945,
      "epoch": 1.0461911125960575,
      "grad_norm": 1.21875,
      "learning_rate": 0.000489618914054139,
      "loss": 4.8827,
      "mean_token_accuracy": 0.2082804411649704,
      "num_tokens": 23043834.0,
      "step": 12525
    },
    {
      "entropy": 5.126025056838989,
      "epoch": 1.0466087537587705,
      "grad_norm": 1.1875,
      "learning_rate": 0.000489609973862412,
      "loss": 4.9746,
      "mean_token_accuracy": 0.19988940060138702,
      "num_tokens": 23052542.0,
      "step": 12530
    },
    {
      "entropy": 5.2754374027252195,
      "epoch": 1.0470263949214835,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004896010299136793,
      "loss": 5.0425,
      "mean_token_accuracy": 0.20176790356636048,
      "num_tokens": 23061438.0,
      "step": 12535
    },
    {
      "entropy": 5.265765762329101,
      "epoch": 1.0474440360841966,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004895920822080972,
      "loss": 5.1375,
      "mean_token_accuracy": 0.18932921439409256,
      "num_tokens": 23071777.0,
      "step": 12540
    },
    {
      "entropy": 5.209864234924316,
      "epoch": 1.0478616772469094,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004895831307458224,
      "loss": 5.0094,
      "mean_token_accuracy": 0.204473876953125,
      "num_tokens": 23080889.0,
      "step": 12545
    },
    {
      "entropy": 5.200679922103882,
      "epoch": 1.0482793184096224,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004895741755270115,
      "loss": 4.9098,
      "mean_token_accuracy": 0.20648445039987565,
      "num_tokens": 23089737.0,
      "step": 12550
    },
    {
      "entropy": 5.1579631805419925,
      "epoch": 1.0486969595723354,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004895652165518215,
      "loss": 4.9715,
      "mean_token_accuracy": 0.20762208849191666,
      "num_tokens": 23098861.0,
      "step": 12555
    },
    {
      "entropy": 5.169628000259399,
      "epoch": 1.0491146007350485,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004895562538204091,
      "loss": 5.0722,
      "mean_token_accuracy": 0.19872246384620668,
      "num_tokens": 23107798.0,
      "step": 12560
    },
    {
      "entropy": 5.209341144561767,
      "epoch": 1.0495322418977615,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004895472873329311,
      "loss": 4.9081,
      "mean_token_accuracy": 0.20617275089025497,
      "num_tokens": 23116698.0,
      "step": 12565
    },
    {
      "entropy": 5.22366852760315,
      "epoch": 1.0499498830604745,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004895383170895447,
      "loss": 5.0802,
      "mean_token_accuracy": 0.19531170576810836,
      "num_tokens": 23125880.0,
      "step": 12570
    },
    {
      "entropy": 5.225423574447632,
      "epoch": 1.0503675242231874,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004895293430904067,
      "loss": 5.1156,
      "mean_token_accuracy": 0.19867901057004927,
      "num_tokens": 23135633.0,
      "step": 12575
    },
    {
      "entropy": 5.1597894668579105,
      "epoch": 1.0507851653859004,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004895203653356745,
      "loss": 4.9464,
      "mean_token_accuracy": 0.20271793156862258,
      "num_tokens": 23144952.0,
      "step": 12580
    },
    {
      "entropy": 5.159872436523438,
      "epoch": 1.0512028065486134,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004895113838255049,
      "loss": 4.973,
      "mean_token_accuracy": 0.20465976893901824,
      "num_tokens": 23154128.0,
      "step": 12585
    },
    {
      "entropy": 5.130985832214355,
      "epoch": 1.0516204477113265,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004895023985600555,
      "loss": 4.886,
      "mean_token_accuracy": 0.21052435636520386,
      "num_tokens": 23162579.0,
      "step": 12590
    },
    {
      "entropy": 5.223650741577148,
      "epoch": 1.0520380888740395,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004894934095394832,
      "loss": 5.0066,
      "mean_token_accuracy": 0.20212711542844772,
      "num_tokens": 23171889.0,
      "step": 12595
    },
    {
      "entropy": 5.127604007720947,
      "epoch": 1.0524557300367525,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004894844167639458,
      "loss": 4.899,
      "mean_token_accuracy": 0.20885633528232575,
      "num_tokens": 23180876.0,
      "step": 12600
    },
    {
      "entropy": 5.083294439315796,
      "epoch": 1.0528733711994653,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004894754202336004,
      "loss": 4.921,
      "mean_token_accuracy": 0.21090091913938522,
      "num_tokens": 23189567.0,
      "step": 12605
    },
    {
      "entropy": 5.27570915222168,
      "epoch": 1.0532910123621784,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004894664199486047,
      "loss": 4.934,
      "mean_token_accuracy": 0.2010876789689064,
      "num_tokens": 23198142.0,
      "step": 12610
    },
    {
      "entropy": 5.247752237319946,
      "epoch": 1.0537086535248914,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004894574159091161,
      "loss": 5.0485,
      "mean_token_accuracy": 0.19577822983264923,
      "num_tokens": 23207143.0,
      "step": 12615
    },
    {
      "entropy": 5.193780040740966,
      "epoch": 1.0541262946876044,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004894484081152922,
      "loss": 4.9491,
      "mean_token_accuracy": 0.19875023663043975,
      "num_tokens": 23215637.0,
      "step": 12620
    },
    {
      "entropy": 5.244784307479859,
      "epoch": 1.0545439358503175,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004894393965672909,
      "loss": 4.9914,
      "mean_token_accuracy": 0.1976357728242874,
      "num_tokens": 23224402.0,
      "step": 12625
    },
    {
      "entropy": 5.406611728668213,
      "epoch": 1.0549615770130305,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004894303812652698,
      "loss": 5.2295,
      "mean_token_accuracy": 0.1821753591299057,
      "num_tokens": 23232669.0,
      "step": 12630
    },
    {
      "entropy": 5.065331983566284,
      "epoch": 1.0553792181757433,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004894213622093867,
      "loss": 4.9542,
      "mean_token_accuracy": 0.19759776890277864,
      "num_tokens": 23242204.0,
      "step": 12635
    },
    {
      "entropy": 5.169221448898315,
      "epoch": 1.0557968593384564,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004894123393997997,
      "loss": 5.0211,
      "mean_token_accuracy": 0.1980602487921715,
      "num_tokens": 23250765.0,
      "step": 12640
    },
    {
      "entropy": 5.180795431137085,
      "epoch": 1.0562145005011694,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004894033128366665,
      "loss": 4.9303,
      "mean_token_accuracy": 0.20086126178503036,
      "num_tokens": 23259986.0,
      "step": 12645
    },
    {
      "entropy": 5.100366401672363,
      "epoch": 1.0566321416638824,
      "grad_norm": 1.25,
      "learning_rate": 0.0004893942825201453,
      "loss": 4.8469,
      "mean_token_accuracy": 0.2096587359905243,
      "num_tokens": 23268599.0,
      "step": 12650
    },
    {
      "entropy": 5.105816507339478,
      "epoch": 1.0570497828265955,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000489385248450394,
      "loss": 4.9329,
      "mean_token_accuracy": 0.20656008422374725,
      "num_tokens": 23277509.0,
      "step": 12655
    },
    {
      "entropy": 5.2050834655761715,
      "epoch": 1.0574674239893085,
      "grad_norm": 1.234375,
      "learning_rate": 0.000489376210627571,
      "loss": 4.9635,
      "mean_token_accuracy": 0.2024905040860176,
      "num_tokens": 23286673.0,
      "step": 12660
    },
    {
      "entropy": 5.2362377643585205,
      "epoch": 1.0578850651520213,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004893671690518343,
      "loss": 4.944,
      "mean_token_accuracy": 0.20659847408533097,
      "num_tokens": 23295775.0,
      "step": 12665
    },
    {
      "entropy": 5.115500164031983,
      "epoch": 1.0583027063147343,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004893581237233423,
      "loss": 4.9666,
      "mean_token_accuracy": 0.20339817255735398,
      "num_tokens": 23305321.0,
      "step": 12670
    },
    {
      "entropy": 5.217950534820557,
      "epoch": 1.0587203474774474,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004893490746422533,
      "loss": 4.9366,
      "mean_token_accuracy": 0.2031608983874321,
      "num_tokens": 23313798.0,
      "step": 12675
    },
    {
      "entropy": 5.098764371871948,
      "epoch": 1.0591379886401604,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004893400218087258,
      "loss": 4.8447,
      "mean_token_accuracy": 0.2129884570837021,
      "num_tokens": 23322308.0,
      "step": 12680
    },
    {
      "entropy": 5.186078596115112,
      "epoch": 1.0595556298028734,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004893309652229183,
      "loss": 5.0557,
      "mean_token_accuracy": 0.19403709471225739,
      "num_tokens": 23331311.0,
      "step": 12685
    },
    {
      "entropy": 5.258445787429809,
      "epoch": 1.0599732709655865,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004893219048849891,
      "loss": 5.1228,
      "mean_token_accuracy": 0.19382873326539993,
      "num_tokens": 23340655.0,
      "step": 12690
    },
    {
      "entropy": 5.27389817237854,
      "epoch": 1.0603909121282993,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004893128407950971,
      "loss": 5.1672,
      "mean_token_accuracy": 0.18957239687442778,
      "num_tokens": 23350999.0,
      "step": 12695
    },
    {
      "entropy": 5.221961832046508,
      "epoch": 1.0608085532910123,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004893037729534007,
      "loss": 4.9855,
      "mean_token_accuracy": 0.2060656651854515,
      "num_tokens": 23359311.0,
      "step": 12700
    },
    {
      "entropy": 5.234956455230713,
      "epoch": 1.0612261944537253,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004892947013600589,
      "loss": 5.0143,
      "mean_token_accuracy": 0.2009303241968155,
      "num_tokens": 23367925.0,
      "step": 12705
    },
    {
      "entropy": 5.217855072021484,
      "epoch": 1.0616438356164384,
      "grad_norm": 1.5,
      "learning_rate": 0.0004892856260152304,
      "loss": 5.1356,
      "mean_token_accuracy": 0.19304483830928804,
      "num_tokens": 23377159.0,
      "step": 12710
    },
    {
      "entropy": 5.389584350585937,
      "epoch": 1.0620614767791514,
      "grad_norm": 1.390625,
      "learning_rate": 0.000489276546919074,
      "loss": 5.1625,
      "mean_token_accuracy": 0.18605200797319413,
      "num_tokens": 23387699.0,
      "step": 12715
    },
    {
      "entropy": 5.231754922866822,
      "epoch": 1.0624791179418644,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004892674640717488,
      "loss": 5.0676,
      "mean_token_accuracy": 0.19894462674856186,
      "num_tokens": 23397520.0,
      "step": 12720
    },
    {
      "entropy": 5.138805818557739,
      "epoch": 1.0628967591045773,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004892583774734138,
      "loss": 4.9893,
      "mean_token_accuracy": 0.1960932970046997,
      "num_tokens": 23406853.0,
      "step": 12725
    },
    {
      "entropy": 5.138727760314941,
      "epoch": 1.0633144002672903,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004892492871242278,
      "loss": 4.9411,
      "mean_token_accuracy": 0.20283298641443254,
      "num_tokens": 23416235.0,
      "step": 12730
    },
    {
      "entropy": 5.188621139526367,
      "epoch": 1.0637320414300033,
      "grad_norm": 1.375,
      "learning_rate": 0.0004892401930243504,
      "loss": 5.0074,
      "mean_token_accuracy": 0.2002476394176483,
      "num_tokens": 23425347.0,
      "step": 12735
    },
    {
      "entropy": 5.318581628799438,
      "epoch": 1.0641496825927164,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004892310951739404,
      "loss": 5.1659,
      "mean_token_accuracy": 0.19354816675186157,
      "num_tokens": 23434905.0,
      "step": 12740
    },
    {
      "entropy": 5.276684093475342,
      "epoch": 1.0645673237554294,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004892219935731571,
      "loss": 4.993,
      "mean_token_accuracy": 0.20004190057516097,
      "num_tokens": 23444495.0,
      "step": 12745
    },
    {
      "entropy": 5.134641838073731,
      "epoch": 1.0649849649181424,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048921288822216,
      "loss": 4.9247,
      "mean_token_accuracy": 0.2023975744843483,
      "num_tokens": 23452224.0,
      "step": 12750
    },
    {
      "entropy": 5.158041381835938,
      "epoch": 1.0654026060808552,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004892037791211086,
      "loss": 5.1137,
      "mean_token_accuracy": 0.19504087865352632,
      "num_tokens": 23461987.0,
      "step": 12755
    },
    {
      "entropy": 5.109542894363403,
      "epoch": 1.0658202472435683,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004891946662701621,
      "loss": 4.9156,
      "mean_token_accuracy": 0.2100577771663666,
      "num_tokens": 23470645.0,
      "step": 12760
    },
    {
      "entropy": 5.246750259399414,
      "epoch": 1.0662378884062813,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004891855496694801,
      "loss": 5.083,
      "mean_token_accuracy": 0.19600992649793625,
      "num_tokens": 23480169.0,
      "step": 12765
    },
    {
      "entropy": 5.2948386669158936,
      "epoch": 1.0666555295689943,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004891764293192222,
      "loss": 5.0224,
      "mean_token_accuracy": 0.19793600142002105,
      "num_tokens": 23488630.0,
      "step": 12770
    },
    {
      "entropy": 5.174128675460816,
      "epoch": 1.0670731707317074,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004891673052195482,
      "loss": 4.8938,
      "mean_token_accuracy": 0.2037169024348259,
      "num_tokens": 23497800.0,
      "step": 12775
    },
    {
      "entropy": 5.182644271850586,
      "epoch": 1.0674908118944204,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004891581773706177,
      "loss": 5.1058,
      "mean_token_accuracy": 0.18771177232265474,
      "num_tokens": 23507400.0,
      "step": 12780
    },
    {
      "entropy": 5.27351131439209,
      "epoch": 1.0679084530571332,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004891490457725906,
      "loss": 5.0982,
      "mean_token_accuracy": 0.18526479303836824,
      "num_tokens": 23516441.0,
      "step": 12785
    },
    {
      "entropy": 5.210230159759521,
      "epoch": 1.0683260942198463,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004891399104256266,
      "loss": 4.9498,
      "mean_token_accuracy": 0.2046144887804985,
      "num_tokens": 23525066.0,
      "step": 12790
    },
    {
      "entropy": 5.139497327804565,
      "epoch": 1.0687437353825593,
      "grad_norm": 1.25,
      "learning_rate": 0.0004891307713298857,
      "loss": 4.9142,
      "mean_token_accuracy": 0.20734929889440537,
      "num_tokens": 23533892.0,
      "step": 12795
    },
    {
      "entropy": 5.192386770248413,
      "epoch": 1.0691613765452723,
      "grad_norm": 1.421875,
      "learning_rate": 0.000489121628485528,
      "loss": 5.0395,
      "mean_token_accuracy": 0.1950652688741684,
      "num_tokens": 23542680.0,
      "step": 12800
    },
    {
      "entropy": 5.1639063358306885,
      "epoch": 1.0695790177079854,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004891124818927133,
      "loss": 4.8527,
      "mean_token_accuracy": 0.2120129019021988,
      "num_tokens": 23550573.0,
      "step": 12805
    },
    {
      "entropy": 5.054487991333008,
      "epoch": 1.0699966588706984,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004891033315516019,
      "loss": 4.9994,
      "mean_token_accuracy": 0.19559263288974763,
      "num_tokens": 23561281.0,
      "step": 12810
    },
    {
      "entropy": 5.14283447265625,
      "epoch": 1.0704143000334112,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004890941774623541,
      "loss": 4.897,
      "mean_token_accuracy": 0.21021474450826644,
      "num_tokens": 23569846.0,
      "step": 12815
    },
    {
      "entropy": 5.192872762680054,
      "epoch": 1.0708319411961242,
      "grad_norm": 1.625,
      "learning_rate": 0.0004890850196251299,
      "loss": 4.9005,
      "mean_token_accuracy": 0.21321782022714614,
      "num_tokens": 23578145.0,
      "step": 12820
    },
    {
      "entropy": 5.139480352401733,
      "epoch": 1.0712495823588373,
      "grad_norm": 1.375,
      "learning_rate": 0.0004890758580400898,
      "loss": 5.0014,
      "mean_token_accuracy": 0.1970194786787033,
      "num_tokens": 23587494.0,
      "step": 12825
    },
    {
      "entropy": 5.280103445053101,
      "epoch": 1.0716672235215503,
      "grad_norm": 1.5,
      "learning_rate": 0.0004890666927073941,
      "loss": 5.0911,
      "mean_token_accuracy": 0.19357855319976808,
      "num_tokens": 23596599.0,
      "step": 12830
    },
    {
      "entropy": 5.175860643386841,
      "epoch": 1.0720848646842633,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004890575236272032,
      "loss": 5.0414,
      "mean_token_accuracy": 0.20384894609451293,
      "num_tokens": 23606170.0,
      "step": 12835
    },
    {
      "entropy": 5.112048387527466,
      "epoch": 1.0725025058469764,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004890483507996778,
      "loss": 4.8552,
      "mean_token_accuracy": 0.21105672121047975,
      "num_tokens": 23614558.0,
      "step": 12840
    },
    {
      "entropy": 5.187344646453857,
      "epoch": 1.0729201470096892,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004890391742249783,
      "loss": 5.0465,
      "mean_token_accuracy": 0.194507959485054,
      "num_tokens": 23624125.0,
      "step": 12845
    },
    {
      "entropy": 5.27942533493042,
      "epoch": 1.0733377881724022,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004890299939032654,
      "loss": 5.0089,
      "mean_token_accuracy": 0.20120906233787536,
      "num_tokens": 23633025.0,
      "step": 12850
    },
    {
      "entropy": 5.128687000274658,
      "epoch": 1.0737554293351153,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004890208098346999,
      "loss": 4.9676,
      "mean_token_accuracy": 0.19719949513673782,
      "num_tokens": 23643387.0,
      "step": 12855
    },
    {
      "entropy": 5.203180170059204,
      "epoch": 1.0741730704978283,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004890116220194425,
      "loss": 5.0318,
      "mean_token_accuracy": 0.1990673780441284,
      "num_tokens": 23653486.0,
      "step": 12860
    },
    {
      "entropy": 5.148255014419556,
      "epoch": 1.0745907116605413,
      "grad_norm": 1.359375,
      "learning_rate": 0.000489002430457654,
      "loss": 4.9315,
      "mean_token_accuracy": 0.200714211165905,
      "num_tokens": 23663483.0,
      "step": 12865
    },
    {
      "entropy": 5.044687795639038,
      "epoch": 1.0750083528232544,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004889932351494954,
      "loss": 4.9734,
      "mean_token_accuracy": 0.2067367449402809,
      "num_tokens": 23672453.0,
      "step": 12870
    },
    {
      "entropy": 5.204858589172363,
      "epoch": 1.0754259939859672,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004889840360951277,
      "loss": 4.9945,
      "mean_token_accuracy": 0.20300368070602418,
      "num_tokens": 23681351.0,
      "step": 12875
    },
    {
      "entropy": 5.234554672241211,
      "epoch": 1.0758436351486802,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004889748332947118,
      "loss": 4.968,
      "mean_token_accuracy": 0.2027100682258606,
      "num_tokens": 23689581.0,
      "step": 12880
    },
    {
      "entropy": 5.219699811935425,
      "epoch": 1.0762612763113932,
      "grad_norm": 1.25,
      "learning_rate": 0.0004889656267484089,
      "loss": 5.065,
      "mean_token_accuracy": 0.1951802298426628,
      "num_tokens": 23699415.0,
      "step": 12885
    },
    {
      "entropy": 5.175791454315186,
      "epoch": 1.0766789174741063,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048895641645638,
      "loss": 4.9931,
      "mean_token_accuracy": 0.20175662785768508,
      "num_tokens": 23708933.0,
      "step": 12890
    },
    {
      "entropy": 5.1675902843475345,
      "epoch": 1.0770965586368193,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004889472024187867,
      "loss": 5.0267,
      "mean_token_accuracy": 0.19115378260612487,
      "num_tokens": 23718423.0,
      "step": 12895
    },
    {
      "entropy": 5.236190509796143,
      "epoch": 1.0775141997995323,
      "grad_norm": 1.34375,
      "learning_rate": 0.00048893798463579,
      "loss": 5.0057,
      "mean_token_accuracy": 0.1984273299574852,
      "num_tokens": 23728131.0,
      "step": 12900
    },
    {
      "entropy": 5.144094848632813,
      "epoch": 1.0779318409622451,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004889287631075513,
      "loss": 4.9887,
      "mean_token_accuracy": 0.20271819233894348,
      "num_tokens": 23737378.0,
      "step": 12905
    },
    {
      "entropy": 5.32651834487915,
      "epoch": 1.0783494821249582,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004889195378342322,
      "loss": 5.0896,
      "mean_token_accuracy": 0.1891043469309807,
      "num_tokens": 23747156.0,
      "step": 12910
    },
    {
      "entropy": 5.2650751113891605,
      "epoch": 1.0787671232876712,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004889103088159941,
      "loss": 5.0324,
      "mean_token_accuracy": 0.197127465903759,
      "num_tokens": 23757090.0,
      "step": 12915
    },
    {
      "entropy": 5.135481071472168,
      "epoch": 1.0791847644503842,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004889010760529986,
      "loss": 5.0371,
      "mean_token_accuracy": 0.19155288934707643,
      "num_tokens": 23766679.0,
      "step": 12920
    },
    {
      "entropy": 5.14391975402832,
      "epoch": 1.0796024056130973,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004888918395454071,
      "loss": 4.9142,
      "mean_token_accuracy": 0.21385714858770372,
      "num_tokens": 23775109.0,
      "step": 12925
    },
    {
      "entropy": 5.1810417652130125,
      "epoch": 1.0800200467758103,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004888825992933817,
      "loss": 5.0041,
      "mean_token_accuracy": 0.20600292682647706,
      "num_tokens": 23784698.0,
      "step": 12930
    },
    {
      "entropy": 5.086528253555298,
      "epoch": 1.0804376879385231,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004888733552970839,
      "loss": 4.9021,
      "mean_token_accuracy": 0.2061927765607834,
      "num_tokens": 23793590.0,
      "step": 12935
    },
    {
      "entropy": 5.155827140808105,
      "epoch": 1.0808553291012362,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004888641075566755,
      "loss": 4.9327,
      "mean_token_accuracy": 0.20301061421632766,
      "num_tokens": 23802421.0,
      "step": 12940
    },
    {
      "entropy": 5.195164442062378,
      "epoch": 1.0812729702639492,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004888548560723184,
      "loss": 5.0513,
      "mean_token_accuracy": 0.19882436394691466,
      "num_tokens": 23811610.0,
      "step": 12945
    },
    {
      "entropy": 5.279752588272094,
      "epoch": 1.0816906114266622,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004888456008441748,
      "loss": 5.0658,
      "mean_token_accuracy": 0.2030701994895935,
      "num_tokens": 23821197.0,
      "step": 12950
    },
    {
      "entropy": 5.264985609054565,
      "epoch": 1.0821082525893753,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004888363418724064,
      "loss": 5.0323,
      "mean_token_accuracy": 0.20127220451831818,
      "num_tokens": 23829757.0,
      "step": 12955
    },
    {
      "entropy": 5.186293315887451,
      "epoch": 1.0825258937520883,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004888270791571756,
      "loss": 5.0071,
      "mean_token_accuracy": 0.19404354244470595,
      "num_tokens": 23839242.0,
      "step": 12960
    },
    {
      "entropy": 5.158607912063599,
      "epoch": 1.082943534914801,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004888178126986444,
      "loss": 4.9403,
      "mean_token_accuracy": 0.2153720274567604,
      "num_tokens": 23848193.0,
      "step": 12965
    },
    {
      "entropy": 5.177860164642334,
      "epoch": 1.0833611760775141,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004888085424969748,
      "loss": 4.927,
      "mean_token_accuracy": 0.20651716887950897,
      "num_tokens": 23857623.0,
      "step": 12970
    },
    {
      "entropy": 5.200423622131348,
      "epoch": 1.0837788172402272,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004887992685523295,
      "loss": 5.0612,
      "mean_token_accuracy": 0.19333951473236083,
      "num_tokens": 23866424.0,
      "step": 12975
    },
    {
      "entropy": 5.201368999481201,
      "epoch": 1.0841964584029402,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004887899908648706,
      "loss": 5.0827,
      "mean_token_accuracy": 0.19666419476270675,
      "num_tokens": 23874855.0,
      "step": 12980
    },
    {
      "entropy": 5.261940145492554,
      "epoch": 1.0846140995656532,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004887807094347606,
      "loss": 5.0509,
      "mean_token_accuracy": 0.20008407086133956,
      "num_tokens": 23884634.0,
      "step": 12985
    },
    {
      "entropy": 5.277802562713623,
      "epoch": 1.0850317407283663,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004887714242621619,
      "loss": 5.0764,
      "mean_token_accuracy": 0.19503322690725328,
      "num_tokens": 23894307.0,
      "step": 12990
    },
    {
      "entropy": 5.204030466079712,
      "epoch": 1.085449381891079,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004887621353472371,
      "loss": 4.9813,
      "mean_token_accuracy": 0.19868019223213196,
      "num_tokens": 23903274.0,
      "step": 12995
    },
    {
      "entropy": 5.175687551498413,
      "epoch": 1.0858670230537921,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004887528426901488,
      "loss": 5.0127,
      "mean_token_accuracy": 0.2007841005921364,
      "num_tokens": 23912787.0,
      "step": 13000
    },
    {
      "entropy": 5.22480583190918,
      "epoch": 1.0862846642165052,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004887435462910597,
      "loss": 5.0505,
      "mean_token_accuracy": 0.19208271503448487,
      "num_tokens": 23921558.0,
      "step": 13005
    },
    {
      "entropy": 5.178594350814819,
      "epoch": 1.0867023053792182,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004887342461501326,
      "loss": 4.9349,
      "mean_token_accuracy": 0.19945605099201202,
      "num_tokens": 23929868.0,
      "step": 13010
    },
    {
      "entropy": 5.168375730514526,
      "epoch": 1.0871199465419312,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004887249422675301,
      "loss": 5.0489,
      "mean_token_accuracy": 0.18642401248216628,
      "num_tokens": 23939206.0,
      "step": 13015
    },
    {
      "entropy": 5.21177487373352,
      "epoch": 1.0875375877046443,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004887156346434154,
      "loss": 4.9599,
      "mean_token_accuracy": 0.20076312124729156,
      "num_tokens": 23948213.0,
      "step": 13020
    },
    {
      "entropy": 5.222215986251831,
      "epoch": 1.087955228867357,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004887063232779513,
      "loss": 5.0004,
      "mean_token_accuracy": 0.19596772342920304,
      "num_tokens": 23957269.0,
      "step": 13025
    },
    {
      "entropy": 5.274312210083008,
      "epoch": 1.08837287003007,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004886970081713006,
      "loss": 5.1216,
      "mean_token_accuracy": 0.1928337723016739,
      "num_tokens": 23967300.0,
      "step": 13030
    },
    {
      "entropy": 5.190327787399292,
      "epoch": 1.0887905111927831,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004886876893236266,
      "loss": 4.9946,
      "mean_token_accuracy": 0.20101721137762069,
      "num_tokens": 23976934.0,
      "step": 13035
    },
    {
      "entropy": 5.163508892059326,
      "epoch": 1.0892081523554962,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004886783667350925,
      "loss": 5.0018,
      "mean_token_accuracy": 0.19977751672267913,
      "num_tokens": 23986119.0,
      "step": 13040
    },
    {
      "entropy": 5.1718346118927006,
      "epoch": 1.0896257935182092,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004886690404058613,
      "loss": 5.0215,
      "mean_token_accuracy": 0.2017348513007164,
      "num_tokens": 23996093.0,
      "step": 13045
    },
    {
      "entropy": 5.310827779769897,
      "epoch": 1.0900434346809222,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004886597103360963,
      "loss": 5.1406,
      "mean_token_accuracy": 0.18590367138385772,
      "num_tokens": 24006585.0,
      "step": 13050
    },
    {
      "entropy": 5.254141139984131,
      "epoch": 1.090461075843635,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004886503765259609,
      "loss": 5.081,
      "mean_token_accuracy": 0.19749283641576768,
      "num_tokens": 24016398.0,
      "step": 13055
    },
    {
      "entropy": 5.2079479694366455,
      "epoch": 1.090878717006348,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004886410389756186,
      "loss": 4.9275,
      "mean_token_accuracy": 0.2002464309334755,
      "num_tokens": 24025988.0,
      "step": 13060
    },
    {
      "entropy": 5.217039680480957,
      "epoch": 1.0912963581690611,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004886316976852326,
      "loss": 5.0123,
      "mean_token_accuracy": 0.20237762182950975,
      "num_tokens": 24036448.0,
      "step": 13065
    },
    {
      "entropy": 5.252889680862427,
      "epoch": 1.0917139993317742,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004886223526549666,
      "loss": 5.1296,
      "mean_token_accuracy": 0.1893712267279625,
      "num_tokens": 24047011.0,
      "step": 13070
    },
    {
      "entropy": 5.154879808425903,
      "epoch": 1.0921316404944872,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004886130038849843,
      "loss": 4.8971,
      "mean_token_accuracy": 0.21107379645109176,
      "num_tokens": 24056332.0,
      "step": 13075
    },
    {
      "entropy": 5.179169082641602,
      "epoch": 1.0925492816572002,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004886036513754491,
      "loss": 5.0795,
      "mean_token_accuracy": 0.19962339997291564,
      "num_tokens": 24065538.0,
      "step": 13080
    },
    {
      "entropy": 5.183998537063599,
      "epoch": 1.092966922819913,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004885942951265251,
      "loss": 5.0047,
      "mean_token_accuracy": 0.19814843088388442,
      "num_tokens": 24074950.0,
      "step": 13085
    },
    {
      "entropy": 5.252826356887818,
      "epoch": 1.093384563982626,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004885849351383756,
      "loss": 5.0522,
      "mean_token_accuracy": 0.20104602575302125,
      "num_tokens": 24084571.0,
      "step": 13090
    },
    {
      "entropy": 5.202955818176269,
      "epoch": 1.093802205145339,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004885755714111649,
      "loss": 5.0735,
      "mean_token_accuracy": 0.19885534346103667,
      "num_tokens": 24093693.0,
      "step": 13095
    },
    {
      "entropy": 5.164794111251831,
      "epoch": 1.0942198463080521,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004885662039450565,
      "loss": 4.9873,
      "mean_token_accuracy": 0.19487125128507615,
      "num_tokens": 24104140.0,
      "step": 13100
    },
    {
      "entropy": 5.238285398483276,
      "epoch": 1.0946374874707652,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004885568327402149,
      "loss": 4.9823,
      "mean_token_accuracy": 0.2018568530678749,
      "num_tokens": 24113107.0,
      "step": 13105
    },
    {
      "entropy": 5.175368976593018,
      "epoch": 1.0950551286334782,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004885474577968038,
      "loss": 5.0135,
      "mean_token_accuracy": 0.20291938334703447,
      "num_tokens": 24122271.0,
      "step": 13110
    },
    {
      "entropy": 5.236986446380615,
      "epoch": 1.095472769796191,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004885380791149874,
      "loss": 5.0732,
      "mean_token_accuracy": 0.1896129220724106,
      "num_tokens": 24132068.0,
      "step": 13115
    },
    {
      "entropy": 5.1817597389221195,
      "epoch": 1.095890410958904,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004885286966949298,
      "loss": 4.9687,
      "mean_token_accuracy": 0.19995463490486146,
      "num_tokens": 24140429.0,
      "step": 13120
    },
    {
      "entropy": 5.216770219802856,
      "epoch": 1.096308052121617,
      "grad_norm": 1.25,
      "learning_rate": 0.0004885193105367955,
      "loss": 5.092,
      "mean_token_accuracy": 0.19164970368146897,
      "num_tokens": 24150563.0,
      "step": 13125
    },
    {
      "entropy": 5.228154468536377,
      "epoch": 1.0967256932843301,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004885099206407486,
      "loss": 4.992,
      "mean_token_accuracy": 0.19577551633119583,
      "num_tokens": 24159059.0,
      "step": 13130
    },
    {
      "entropy": 5.285415363311768,
      "epoch": 1.0971433344470432,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004885005270069535,
      "loss": 5.1177,
      "mean_token_accuracy": 0.19171402603387833,
      "num_tokens": 24168858.0,
      "step": 13135
    },
    {
      "entropy": 5.2112569332122805,
      "epoch": 1.0975609756097562,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004884911296355746,
      "loss": 5.0352,
      "mean_token_accuracy": 0.1958905965089798,
      "num_tokens": 24177573.0,
      "step": 13140
    },
    {
      "entropy": 5.187708902359009,
      "epoch": 1.097978616772469,
      "grad_norm": 1.375,
      "learning_rate": 0.0004884817285267766,
      "loss": 4.9601,
      "mean_token_accuracy": 0.20276569873094558,
      "num_tokens": 24186832.0,
      "step": 13145
    },
    {
      "entropy": 5.1466821193695065,
      "epoch": 1.098396257935182,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004884723236807239,
      "loss": 4.9818,
      "mean_token_accuracy": 0.20125413984060286,
      "num_tokens": 24195281.0,
      "step": 13150
    },
    {
      "entropy": 5.184046030044556,
      "epoch": 1.098813899097895,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004884629150975814,
      "loss": 5.0046,
      "mean_token_accuracy": 0.20098939687013626,
      "num_tokens": 24204499.0,
      "step": 13155
    },
    {
      "entropy": 5.10465259552002,
      "epoch": 1.099231540260608,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004884535027775135,
      "loss": 4.8706,
      "mean_token_accuracy": 0.21302092373371123,
      "num_tokens": 24213418.0,
      "step": 13160
    },
    {
      "entropy": 5.065260648727417,
      "epoch": 1.0996491814233211,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004884440867206851,
      "loss": 4.8928,
      "mean_token_accuracy": 0.207216976583004,
      "num_tokens": 24222834.0,
      "step": 13165
    },
    {
      "entropy": 5.159618711471557,
      "epoch": 1.1000668225860342,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000488434666927261,
      "loss": 4.9829,
      "mean_token_accuracy": 0.199624764919281,
      "num_tokens": 24231821.0,
      "step": 13170
    },
    {
      "entropy": 5.245429563522339,
      "epoch": 1.100484463748747,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004884252433974062,
      "loss": 4.9918,
      "mean_token_accuracy": 0.19742787033319473,
      "num_tokens": 24240694.0,
      "step": 13175
    },
    {
      "entropy": 5.217895793914795,
      "epoch": 1.10090210491146,
      "grad_norm": 1.125,
      "learning_rate": 0.0004884158161312856,
      "loss": 5.0251,
      "mean_token_accuracy": 0.1998318240046501,
      "num_tokens": 24249927.0,
      "step": 13180
    },
    {
      "entropy": 5.187979793548584,
      "epoch": 1.101319746074173,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004884063851290642,
      "loss": 4.9919,
      "mean_token_accuracy": 0.20024751722812653,
      "num_tokens": 24259174.0,
      "step": 13185
    },
    {
      "entropy": 5.176235008239746,
      "epoch": 1.101737387236886,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004883969503909073,
      "loss": 5.0226,
      "mean_token_accuracy": 0.20023259967565538,
      "num_tokens": 24268381.0,
      "step": 13190
    },
    {
      "entropy": 5.236040735244751,
      "epoch": 1.1021550283995991,
      "grad_norm": 1.25,
      "learning_rate": 0.0004883875119169798,
      "loss": 5.0888,
      "mean_token_accuracy": 0.19630985110998153,
      "num_tokens": 24277916.0,
      "step": 13195
    },
    {
      "entropy": 5.286180305480957,
      "epoch": 1.1025726695623121,
      "grad_norm": 1.21875,
      "learning_rate": 0.000488378069707447,
      "loss": 5.0959,
      "mean_token_accuracy": 0.1945745289325714,
      "num_tokens": 24287530.0,
      "step": 13200
    },
    {
      "entropy": 5.276157283782959,
      "epoch": 1.102990310725025,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004883686237624744,
      "loss": 5.2095,
      "mean_token_accuracy": 0.1858066737651825,
      "num_tokens": 24298239.0,
      "step": 13205
    },
    {
      "entropy": 5.301026058197022,
      "epoch": 1.103407951887738,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004883591740822271,
      "loss": 5.0221,
      "mean_token_accuracy": 0.1928513154387474,
      "num_tokens": 24307001.0,
      "step": 13210
    },
    {
      "entropy": 5.232964420318604,
      "epoch": 1.103825593050451,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004883497206668707,
      "loss": 4.9699,
      "mean_token_accuracy": 0.20553400367498398,
      "num_tokens": 24315995.0,
      "step": 13215
    },
    {
      "entropy": 5.134984731674194,
      "epoch": 1.104243234213164,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004883402635165705,
      "loss": 4.9165,
      "mean_token_accuracy": 0.2076024070382118,
      "num_tokens": 24324421.0,
      "step": 13220
    },
    {
      "entropy": 5.152343511581421,
      "epoch": 1.104660875375877,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004883308026314923,
      "loss": 5.053,
      "mean_token_accuracy": 0.20021531730890274,
      "num_tokens": 24333715.0,
      "step": 13225
    },
    {
      "entropy": 5.207974481582641,
      "epoch": 1.1050785165385901,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004883213380118016,
      "loss": 4.9598,
      "mean_token_accuracy": 0.20934053659439086,
      "num_tokens": 24342085.0,
      "step": 13230
    },
    {
      "entropy": 5.258758544921875,
      "epoch": 1.105496157701303,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004883118696576642,
      "loss": 5.0656,
      "mean_token_accuracy": 0.1936531275510788,
      "num_tokens": 24350906.0,
      "step": 13235
    },
    {
      "entropy": 5.142571640014649,
      "epoch": 1.105913798864016,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004883023975692457,
      "loss": 4.9211,
      "mean_token_accuracy": 0.20058528929948807,
      "num_tokens": 24359948.0,
      "step": 13240
    },
    {
      "entropy": 5.074871969223023,
      "epoch": 1.106331440026729,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004882929217467118,
      "loss": 4.9169,
      "mean_token_accuracy": 0.20703262239694595,
      "num_tokens": 24369440.0,
      "step": 13245
    },
    {
      "entropy": 5.169613075256348,
      "epoch": 1.106749081189442,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004882834421902287,
      "loss": 5.018,
      "mean_token_accuracy": 0.20072455555200577,
      "num_tokens": 24378442.0,
      "step": 13250
    },
    {
      "entropy": 5.137157392501831,
      "epoch": 1.107166722352155,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00048827395889996226,
      "loss": 4.9972,
      "mean_token_accuracy": 0.20076481699943544,
      "num_tokens": 24387750.0,
      "step": 13255
    },
    {
      "entropy": 5.145231866836548,
      "epoch": 1.107584363514868,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004882644718760784,
      "loss": 4.8132,
      "mean_token_accuracy": 0.21493207812309265,
      "num_tokens": 24396380.0,
      "step": 13260
    },
    {
      "entropy": 5.104736042022705,
      "epoch": 1.108002004677581,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00048825498111874325,
      "loss": 4.8825,
      "mean_token_accuracy": 0.20871036350727082,
      "num_tokens": 24405135.0,
      "step": 13265
    },
    {
      "entropy": 5.128010702133179,
      "epoch": 1.108419645840294,
      "grad_norm": 1.359375,
      "learning_rate": 0.00048824548662812294,
      "loss": 5.0089,
      "mean_token_accuracy": 0.20094665437936782,
      "num_tokens": 24413517.0,
      "step": 13270
    },
    {
      "entropy": 5.217051935195923,
      "epoch": 1.108837287003007,
      "grad_norm": 1.75,
      "learning_rate": 0.00048823598840438384,
      "loss": 5.0044,
      "mean_token_accuracy": 0.2006419599056244,
      "num_tokens": 24422170.0,
      "step": 13275
    },
    {
      "entropy": 5.322580623626709,
      "epoch": 1.10925492816572,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00048822648644769194,
      "loss": 5.1707,
      "mean_token_accuracy": 0.1880909264087677,
      "num_tokens": 24431083.0,
      "step": 13280
    },
    {
      "entropy": 5.2469717502594,
      "epoch": 1.109672569328433,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004882169807582138,
      "loss": 5.0397,
      "mean_token_accuracy": 0.1949939027428627,
      "num_tokens": 24440582.0,
      "step": 13285
    },
    {
      "entropy": 5.233287811279297,
      "epoch": 1.110090210491146,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004882074713361157,
      "loss": 4.9942,
      "mean_token_accuracy": 0.20562419891357422,
      "num_tokens": 24450047.0,
      "step": 13290
    },
    {
      "entropy": 5.09965968132019,
      "epoch": 1.110507851653859,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004881979581815642,
      "loss": 4.8652,
      "mean_token_accuracy": 0.20652866810560228,
      "num_tokens": 24457955.0,
      "step": 13295
    },
    {
      "entropy": 5.067931079864502,
      "epoch": 1.110925492816572,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004881884412947259,
      "loss": 4.8558,
      "mean_token_accuracy": 0.20581793338060378,
      "num_tokens": 24467394.0,
      "step": 13300
    },
    {
      "entropy": 5.155812311172485,
      "epoch": 1.111343133979285,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00048817892067576726,
      "loss": 4.912,
      "mean_token_accuracy": 0.20720148086547852,
      "num_tokens": 24476386.0,
      "step": 13305
    },
    {
      "entropy": 5.136502599716186,
      "epoch": 1.111760775141998,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000488169396324855,
      "loss": 4.9687,
      "mean_token_accuracy": 0.20548605769872666,
      "num_tokens": 24484800.0,
      "step": 13310
    },
    {
      "entropy": 5.181780529022217,
      "epoch": 1.112178416304711,
      "grad_norm": 1.453125,
      "learning_rate": 0.00048815986824215585,
      "loss": 4.9422,
      "mean_token_accuracy": 0.20887642502784728,
      "num_tokens": 24494252.0,
      "step": 13315
    },
    {
      "entropy": 5.156550407409668,
      "epoch": 1.112596057467424,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004881503364278366,
      "loss": 4.9919,
      "mean_token_accuracy": 0.20255128741264344,
      "num_tokens": 24503315.0,
      "step": 13320
    },
    {
      "entropy": 5.203254699707031,
      "epoch": 1.1130136986301369,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004881408008820642,
      "loss": 4.9715,
      "mean_token_accuracy": 0.2068322017788887,
      "num_tokens": 24512660.0,
      "step": 13325
    },
    {
      "entropy": 5.203128051757813,
      "epoch": 1.11343133979285,
      "grad_norm": 1.328125,
      "learning_rate": 0.00048813126160500555,
      "loss": 4.9925,
      "mean_token_accuracy": 0.19681662172079087,
      "num_tokens": 24521712.0,
      "step": 13330
    },
    {
      "entropy": 5.090267896652222,
      "epoch": 1.113848980955563,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004881217185968276,
      "loss": 4.9341,
      "mean_token_accuracy": 0.2146473318338394,
      "num_tokens": 24529851.0,
      "step": 13335
    },
    {
      "entropy": 5.19201169013977,
      "epoch": 1.114266622118276,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004881121718576974,
      "loss": 5.0116,
      "mean_token_accuracy": 0.19630802124738694,
      "num_tokens": 24538425.0,
      "step": 13340
    },
    {
      "entropy": 5.133922338485718,
      "epoch": 1.114684263280989,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00048810262138778213,
      "loss": 5.0932,
      "mean_token_accuracy": 0.19665394127368926,
      "num_tokens": 24547477.0,
      "step": 13345
    },
    {
      "entropy": 5.237164783477783,
      "epoch": 1.115101904443702,
      "grad_norm": 1.328125,
      "learning_rate": 0.00048809306718724886,
      "loss": 5.0455,
      "mean_token_accuracy": 0.19071023166179657,
      "num_tokens": 24556230.0,
      "step": 13350
    },
    {
      "entropy": 5.226452589035034,
      "epoch": 1.115519545606415,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00048808350925626514,
      "loss": 4.9915,
      "mean_token_accuracy": 0.19961708784103394,
      "num_tokens": 24565021.0,
      "step": 13355
    },
    {
      "entropy": 5.228827905654907,
      "epoch": 1.115937186769128,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00048807394759499797,
      "loss": 4.9884,
      "mean_token_accuracy": 0.2000695064663887,
      "num_tokens": 24575012.0,
      "step": 13360
    },
    {
      "entropy": 5.28745846748352,
      "epoch": 1.116354827931841,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004880643822036149,
      "loss": 5.1188,
      "mean_token_accuracy": 0.19034745842218398,
      "num_tokens": 24584267.0,
      "step": 13365
    },
    {
      "entropy": 5.21623649597168,
      "epoch": 1.116772469094554,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00048805481308228333,
      "loss": 5.0301,
      "mean_token_accuracy": 0.19698047488927842,
      "num_tokens": 24594299.0,
      "step": 13370
    },
    {
      "entropy": 5.2318719863891605,
      "epoch": 1.117190110257267,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004880452402311709,
      "loss": 5.016,
      "mean_token_accuracy": 0.19894906878471375,
      "num_tokens": 24603717.0,
      "step": 13375
    },
    {
      "entropy": 5.253253364562989,
      "epoch": 1.11760775141998,
      "grad_norm": 1.578125,
      "learning_rate": 0.00048803566365044493,
      "loss": 5.0688,
      "mean_token_accuracy": 0.1950701132416725,
      "num_tokens": 24613141.0,
      "step": 13380
    },
    {
      "entropy": 5.20740909576416,
      "epoch": 1.1180253925826928,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004880260833402734,
      "loss": 5.009,
      "mean_token_accuracy": 0.20294474065303802,
      "num_tokens": 24621837.0,
      "step": 13385
    },
    {
      "entropy": 5.2285728931427,
      "epoch": 1.1184430337454059,
      "grad_norm": 2.625,
      "learning_rate": 0.00048801649930082375,
      "loss": 5.0604,
      "mean_token_accuracy": 0.19227154850959777,
      "num_tokens": 24630657.0,
      "step": 13390
    },
    {
      "entropy": 5.344229650497437,
      "epoch": 1.118860674908119,
      "grad_norm": 1.578125,
      "learning_rate": 0.000488006911532264,
      "loss": 5.1381,
      "mean_token_accuracy": 0.18907244503498077,
      "num_tokens": 24639929.0,
      "step": 13395
    },
    {
      "entropy": 5.2597150802612305,
      "epoch": 1.119278316070832,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004879973200347617,
      "loss": 5.0718,
      "mean_token_accuracy": 0.20136698931455613,
      "num_tokens": 24648537.0,
      "step": 13400
    },
    {
      "entropy": 5.18208498954773,
      "epoch": 1.119695957233545,
      "grad_norm": 1.296875,
      "learning_rate": 0.00048798772480848504,
      "loss": 5.0249,
      "mean_token_accuracy": 0.20535873472690583,
      "num_tokens": 24656809.0,
      "step": 13405
    },
    {
      "entropy": 5.194003391265869,
      "epoch": 1.120113598396258,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00048797812585360187,
      "loss": 5.0118,
      "mean_token_accuracy": 0.1954981118440628,
      "num_tokens": 24665523.0,
      "step": 13410
    },
    {
      "entropy": 5.236606502532959,
      "epoch": 1.120531239558971,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00048796852317028024,
      "loss": 4.9587,
      "mean_token_accuracy": 0.20104067474603654,
      "num_tokens": 24675618.0,
      "step": 13415
    },
    {
      "entropy": 5.212696123123169,
      "epoch": 1.1209488807216839,
      "grad_norm": 1.5,
      "learning_rate": 0.0004879589167586882,
      "loss": 5.0222,
      "mean_token_accuracy": 0.2022966206073761,
      "num_tokens": 24684922.0,
      "step": 13420
    },
    {
      "entropy": 5.165245246887207,
      "epoch": 1.121366521884397,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048794930661899414,
      "loss": 5.0469,
      "mean_token_accuracy": 0.20175616294145585,
      "num_tokens": 24693791.0,
      "step": 13425
    },
    {
      "entropy": 5.135347175598144,
      "epoch": 1.12178416304711,
      "grad_norm": 1.34375,
      "learning_rate": 0.00048793969275136593,
      "loss": 4.9298,
      "mean_token_accuracy": 0.20342667251825333,
      "num_tokens": 24702661.0,
      "step": 13430
    },
    {
      "entropy": 5.202179670333862,
      "epoch": 1.122201804209823,
      "grad_norm": 1.25,
      "learning_rate": 0.0004879300751559722,
      "loss": 5.0203,
      "mean_token_accuracy": 0.19860461354255676,
      "num_tokens": 24711019.0,
      "step": 13435
    },
    {
      "entropy": 5.279245805740357,
      "epoch": 1.122619445372536,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004879204538329812,
      "loss": 5.0518,
      "mean_token_accuracy": 0.19145938158035278,
      "num_tokens": 24720456.0,
      "step": 13440
    },
    {
      "entropy": 5.213824796676636,
      "epoch": 1.1230370865352488,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004879108287825614,
      "loss": 5.0027,
      "mean_token_accuracy": 0.2010293111205101,
      "num_tokens": 24729395.0,
      "step": 13445
    },
    {
      "entropy": 5.190695714950562,
      "epoch": 1.1234547276979618,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00048790120000488115,
      "loss": 4.9861,
      "mean_token_accuracy": 0.2052915558218956,
      "num_tokens": 24738543.0,
      "step": 13450
    },
    {
      "entropy": 5.202814674377441,
      "epoch": 1.1238723688606749,
      "grad_norm": 1.625,
      "learning_rate": 0.0004878915675001092,
      "loss": 4.9864,
      "mean_token_accuracy": 0.19998397082090377,
      "num_tokens": 24747860.0,
      "step": 13455
    },
    {
      "entropy": 5.113884353637696,
      "epoch": 1.124290010023388,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004878819312684141,
      "loss": 4.9432,
      "mean_token_accuracy": 0.19832088202238082,
      "num_tokens": 24757457.0,
      "step": 13460
    },
    {
      "entropy": 5.3003686428070065,
      "epoch": 1.124707651186101,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004878722913099645,
      "loss": 5.0038,
      "mean_token_accuracy": 0.19819665402173997,
      "num_tokens": 24766844.0,
      "step": 13465
    },
    {
      "entropy": 5.2475510597229,
      "epoch": 1.125125292348814,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004878626476249293,
      "loss": 5.0647,
      "mean_token_accuracy": 0.19061653614044188,
      "num_tokens": 24776041.0,
      "step": 13470
    },
    {
      "entropy": 5.2006142139434814,
      "epoch": 1.125542933511527,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004878530002134772,
      "loss": 5.0186,
      "mean_token_accuracy": 0.20658609867095948,
      "num_tokens": 24785515.0,
      "step": 13475
    },
    {
      "entropy": 5.312012100219727,
      "epoch": 1.1259605746742398,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004878433490757771,
      "loss": 5.0657,
      "mean_token_accuracy": 0.19991444051265717,
      "num_tokens": 24795251.0,
      "step": 13480
    },
    {
      "entropy": 5.149464845657349,
      "epoch": 1.1263782158369529,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004878336942119979,
      "loss": 4.9705,
      "mean_token_accuracy": 0.2093373119831085,
      "num_tokens": 24804672.0,
      "step": 13485
    },
    {
      "entropy": 5.217998743057251,
      "epoch": 1.126795856999666,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00048782403562230874,
      "loss": 5.0007,
      "mean_token_accuracy": 0.20051393806934356,
      "num_tokens": 24813520.0,
      "step": 13490
    },
    {
      "entropy": 5.176961612701416,
      "epoch": 1.127213498162379,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004878143733068787,
      "loss": 4.878,
      "mean_token_accuracy": 0.20750026404857635,
      "num_tokens": 24821368.0,
      "step": 13495
    },
    {
      "entropy": 5.139022588729858,
      "epoch": 1.127631139325092,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004878047072658769,
      "loss": 4.9626,
      "mean_token_accuracy": 0.20105356723070145,
      "num_tokens": 24830071.0,
      "step": 13500
    },
    {
      "entropy": 5.093559789657593,
      "epoch": 1.1280487804878048,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004877950374994725,
      "loss": 4.8726,
      "mean_token_accuracy": 0.20642393827438354,
      "num_tokens": 24838982.0,
      "step": 13505
    },
    {
      "entropy": 5.230965995788575,
      "epoch": 1.1284664216505178,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004877853640078349,
      "loss": 5.0718,
      "mean_token_accuracy": 0.18782105594873427,
      "num_tokens": 24848313.0,
      "step": 13510
    },
    {
      "entropy": 5.175296831130981,
      "epoch": 1.1288840628132308,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004877756867911333,
      "loss": 5.0396,
      "mean_token_accuracy": 0.20018076598644258,
      "num_tokens": 24857477.0,
      "step": 13515
    },
    {
      "entropy": 5.202656555175781,
      "epoch": 1.1293017039759439,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004877660058495372,
      "loss": 4.9885,
      "mean_token_accuracy": 0.20033082515001296,
      "num_tokens": 24866561.0,
      "step": 13520
    },
    {
      "entropy": 5.238004541397094,
      "epoch": 1.129719345138657,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004877563211832161,
      "loss": 5.064,
      "mean_token_accuracy": 0.188402858376503,
      "num_tokens": 24876113.0,
      "step": 13525
    },
    {
      "entropy": 5.311277866363525,
      "epoch": 1.13013698630137,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00048774663279233947,
      "loss": 5.1302,
      "mean_token_accuracy": 0.1879250168800354,
      "num_tokens": 24885974.0,
      "step": 13530
    },
    {
      "entropy": 5.219123697280883,
      "epoch": 1.130554627464083,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00048773694067707693,
      "loss": 4.9825,
      "mean_token_accuracy": 0.20418574512004853,
      "num_tokens": 24895046.0,
      "step": 13535
    },
    {
      "entropy": 5.22024188041687,
      "epoch": 1.1309722686267958,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00048772724483759825,
      "loss": 5.1131,
      "mean_token_accuracy": 0.19033268839120865,
      "num_tokens": 24903315.0,
      "step": 13540
    },
    {
      "entropy": 5.247597789764404,
      "epoch": 1.1313899097895088,
      "grad_norm": 1.265625,
      "learning_rate": 0.000487717545274073,
      "loss": 5.115,
      "mean_token_accuracy": 0.19516629874706268,
      "num_tokens": 24912556.0,
      "step": 13545
    },
    {
      "entropy": 5.303421926498413,
      "epoch": 1.1318075509522219,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004877078419866711,
      "loss": 5.0241,
      "mean_token_accuracy": 0.2046926960349083,
      "num_tokens": 24921854.0,
      "step": 13550
    },
    {
      "entropy": 5.191117525100708,
      "epoch": 1.1322251921149349,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00048769813497556236,
      "loss": 4.9639,
      "mean_token_accuracy": 0.20821459889411925,
      "num_tokens": 24930691.0,
      "step": 13555
    },
    {
      "entropy": 5.197044944763183,
      "epoch": 1.132642833277648,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004876884242409168,
      "loss": 5.0425,
      "mean_token_accuracy": 0.19681573212146758,
      "num_tokens": 24940573.0,
      "step": 13560
    },
    {
      "entropy": 5.215366983413697,
      "epoch": 1.1330604744403607,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00048767870978290434,
      "loss": 4.9401,
      "mean_token_accuracy": 0.19953376799821854,
      "num_tokens": 24950902.0,
      "step": 13565
    },
    {
      "entropy": 5.132641839981079,
      "epoch": 1.1334781156030738,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004876689916016951,
      "loss": 4.9655,
      "mean_token_accuracy": 0.2007038787007332,
      "num_tokens": 24960055.0,
      "step": 13570
    },
    {
      "entropy": 5.174301195144653,
      "epoch": 1.1338957567657868,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004876592696974591,
      "loss": 4.9964,
      "mean_token_accuracy": 0.2013462096452713,
      "num_tokens": 24969896.0,
      "step": 13575
    },
    {
      "entropy": 5.237551498413086,
      "epoch": 1.1343133979284998,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004876495440703667,
      "loss": 4.9535,
      "mean_token_accuracy": 0.20290314853191377,
      "num_tokens": 24978467.0,
      "step": 13580
    },
    {
      "entropy": 5.179347801208496,
      "epoch": 1.1347310390912129,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004876398147205879,
      "loss": 5.0125,
      "mean_token_accuracy": 0.20421649068593978,
      "num_tokens": 24987038.0,
      "step": 13585
    },
    {
      "entropy": 5.285837888717651,
      "epoch": 1.135148680253926,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004876300816482933,
      "loss": 5.0743,
      "mean_token_accuracy": 0.19636330753564835,
      "num_tokens": 24997228.0,
      "step": 13590
    },
    {
      "entropy": 5.185794830322266,
      "epoch": 1.135566321416639,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004876203448536532,
      "loss": 4.9551,
      "mean_token_accuracy": 0.19918005615472795,
      "num_tokens": 25005784.0,
      "step": 13595
    },
    {
      "entropy": 5.167996978759765,
      "epoch": 1.1359839625793517,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004876106043368381,
      "loss": 5.0514,
      "mean_token_accuracy": 0.1936498150229454,
      "num_tokens": 25015863.0,
      "step": 13600
    },
    {
      "entropy": 5.063472938537598,
      "epoch": 1.1364016037420648,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004876008600980183,
      "loss": 4.8779,
      "mean_token_accuracy": 0.2128603368997574,
      "num_tokens": 25025391.0,
      "step": 13605
    },
    {
      "entropy": 5.242287492752075,
      "epoch": 1.1368192449047778,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00048759111213736455,
      "loss": 5.1073,
      "mean_token_accuracy": 0.19428230822086334,
      "num_tokens": 25035387.0,
      "step": 13610
    },
    {
      "entropy": 5.158678960800171,
      "epoch": 1.1372368860674908,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004875813604550476,
      "loss": 4.8359,
      "mean_token_accuracy": 0.2054018959403038,
      "num_tokens": 25044485.0,
      "step": 13615
    },
    {
      "entropy": 5.0899741649627686,
      "epoch": 1.1376545272302039,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004875716050512379,
      "loss": 4.9772,
      "mean_token_accuracy": 0.20481752306222917,
      "num_tokens": 25053079.0,
      "step": 13620
    },
    {
      "entropy": 5.271360778808594,
      "epoch": 1.1380721683929167,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004875618459261065,
      "loss": 5.0867,
      "mean_token_accuracy": 0.19302895814180374,
      "num_tokens": 25063262.0,
      "step": 13625
    },
    {
      "entropy": 5.21492600440979,
      "epoch": 1.1384898095556297,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00048755208307982414,
      "loss": 4.9688,
      "mean_token_accuracy": 0.20691935271024703,
      "num_tokens": 25071720.0,
      "step": 13630
    },
    {
      "entropy": 5.098860597610473,
      "epoch": 1.1389074507183428,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004875423165125616,
      "loss": 4.9911,
      "mean_token_accuracy": 0.20724838972091675,
      "num_tokens": 25080460.0,
      "step": 13635
    },
    {
      "entropy": 5.222128677368164,
      "epoch": 1.1393250918810558,
      "grad_norm": 1.28125,
      "learning_rate": 0.00048753254622449006,
      "loss": 4.9898,
      "mean_token_accuracy": 0.20312332212924958,
      "num_tokens": 25089162.0,
      "step": 13640
    },
    {
      "entropy": 5.146056079864502,
      "epoch": 1.1397427330437688,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004875227722157804,
      "loss": 4.8742,
      "mean_token_accuracy": 0.2038775071501732,
      "num_tokens": 25098234.0,
      "step": 13645
    },
    {
      "entropy": 5.094821119308472,
      "epoch": 1.1401603742064819,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004875129944866039,
      "loss": 4.958,
      "mean_token_accuracy": 0.2000128448009491,
      "num_tokens": 25107832.0,
      "step": 13650
    },
    {
      "entropy": 5.252669477462769,
      "epoch": 1.140578015369195,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004875032130371316,
      "loss": 4.9735,
      "mean_token_accuracy": 0.20208201855421065,
      "num_tokens": 25116118.0,
      "step": 13655
    },
    {
      "entropy": 5.109338045120239,
      "epoch": 1.1409956565319077,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004874934278675347,
      "loss": 4.8067,
      "mean_token_accuracy": 0.212659615278244,
      "num_tokens": 25125025.0,
      "step": 13660
    },
    {
      "entropy": 5.126728773117065,
      "epoch": 1.1414132976946207,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004874836389779846,
      "loss": 5.0878,
      "mean_token_accuracy": 0.19592613130807876,
      "num_tokens": 25135763.0,
      "step": 13665
    },
    {
      "entropy": 5.300793552398682,
      "epoch": 1.1418309388573338,
      "grad_norm": 1.625,
      "learning_rate": 0.00048747384636865267,
      "loss": 5.1144,
      "mean_token_accuracy": 0.19238030314445495,
      "num_tokens": 25145510.0,
      "step": 13670
    },
    {
      "entropy": 5.25956449508667,
      "epoch": 1.1422485800200468,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004874640500397103,
      "loss": 4.9445,
      "mean_token_accuracy": 0.1995074450969696,
      "num_tokens": 25153910.0,
      "step": 13675
    },
    {
      "entropy": 5.162671947479248,
      "epoch": 1.1426662211827598,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004874542499913289,
      "loss": 4.9919,
      "mean_token_accuracy": 0.19615438580513,
      "num_tokens": 25163357.0,
      "step": 13680
    },
    {
      "entropy": 5.188836622238159,
      "epoch": 1.1430838623454727,
      "grad_norm": 1.328125,
      "learning_rate": 0.00048744444622368015,
      "loss": 4.9961,
      "mean_token_accuracy": 0.2027961015701294,
      "num_tokens": 25172423.0,
      "step": 13685
    },
    {
      "entropy": 5.231748580932617,
      "epoch": 1.1435015035081857,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048743463873693576,
      "loss": 4.932,
      "mean_token_accuracy": 0.199450021982193,
      "num_tokens": 25181508.0,
      "step": 13690
    },
    {
      "entropy": 5.183304500579834,
      "epoch": 1.1439191446708987,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004874248275312673,
      "loss": 5.059,
      "mean_token_accuracy": 0.2030150055885315,
      "num_tokens": 25190741.0,
      "step": 13695
    },
    {
      "entropy": 5.26699481010437,
      "epoch": 1.1443367858336118,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00048741501260684644,
      "loss": 5.1263,
      "mean_token_accuracy": 0.19087373316287995,
      "num_tokens": 25201288.0,
      "step": 13700
    },
    {
      "entropy": 5.2198995590209964,
      "epoch": 1.1447544269963248,
      "grad_norm": 1.515625,
      "learning_rate": 0.00048740519396384515,
      "loss": 4.9857,
      "mean_token_accuracy": 0.20888454765081405,
      "num_tokens": 25210111.0,
      "step": 13705
    },
    {
      "entropy": 5.207336187362671,
      "epoch": 1.1451720681590378,
      "grad_norm": 1.21875,
      "learning_rate": 0.00048739537160243524,
      "loss": 5.0204,
      "mean_token_accuracy": 0.20029985755681992,
      "num_tokens": 25220061.0,
      "step": 13710
    },
    {
      "entropy": 5.237885808944702,
      "epoch": 1.1455897093217509,
      "grad_norm": 1.25,
      "learning_rate": 0.00048738554552278874,
      "loss": 5.0202,
      "mean_token_accuracy": 0.20138348489999772,
      "num_tokens": 25229336.0,
      "step": 13715
    },
    {
      "entropy": 5.159564399719239,
      "epoch": 1.1460073504844637,
      "grad_norm": 1.25,
      "learning_rate": 0.00048737571572507753,
      "loss": 4.9465,
      "mean_token_accuracy": 0.208954918384552,
      "num_tokens": 25238539.0,
      "step": 13720
    },
    {
      "entropy": 5.201770877838134,
      "epoch": 1.1464249916471767,
      "grad_norm": 1.25,
      "learning_rate": 0.00048736588220947384,
      "loss": 4.997,
      "mean_token_accuracy": 0.20012432485818862,
      "num_tokens": 25247880.0,
      "step": 13725
    },
    {
      "entropy": 5.22841248512268,
      "epoch": 1.1468426328098897,
      "grad_norm": 1.28125,
      "learning_rate": 0.00048735604497614974,
      "loss": 5.0377,
      "mean_token_accuracy": 0.1979969009757042,
      "num_tokens": 25257295.0,
      "step": 13730
    },
    {
      "entropy": 5.194210577011108,
      "epoch": 1.1472602739726028,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004873462040252775,
      "loss": 5.0343,
      "mean_token_accuracy": 0.1985902041196823,
      "num_tokens": 25266784.0,
      "step": 13735
    },
    {
      "entropy": 5.263298654556275,
      "epoch": 1.1476779151353158,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004873363593570293,
      "loss": 5.0806,
      "mean_token_accuracy": 0.19196487516164779,
      "num_tokens": 25276910.0,
      "step": 13740
    },
    {
      "entropy": 5.256287574768066,
      "epoch": 1.1480955562980286,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004873265109715776,
      "loss": 4.9754,
      "mean_token_accuracy": 0.2044925570487976,
      "num_tokens": 25285776.0,
      "step": 13745
    },
    {
      "entropy": 5.188464736938476,
      "epoch": 1.1485131974607417,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004873166588690947,
      "loss": 4.9509,
      "mean_token_accuracy": 0.19964724332094191,
      "num_tokens": 25294319.0,
      "step": 13750
    },
    {
      "entropy": 5.099911308288574,
      "epoch": 1.1489308386234547,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004873068030497532,
      "loss": 4.9538,
      "mean_token_accuracy": 0.2054399609565735,
      "num_tokens": 25304396.0,
      "step": 13755
    },
    {
      "entropy": 5.143560552597046,
      "epoch": 1.1493484797861677,
      "grad_norm": 1.296875,
      "learning_rate": 0.00048729694351372545,
      "loss": 4.9188,
      "mean_token_accuracy": 0.20863095670938492,
      "num_tokens": 25313095.0,
      "step": 13760
    },
    {
      "entropy": 5.229639768600464,
      "epoch": 1.1497661209488808,
      "grad_norm": 1.375,
      "learning_rate": 0.0004872870802611842,
      "loss": 5.1098,
      "mean_token_accuracy": 0.20023561716079713,
      "num_tokens": 25322948.0,
      "step": 13765
    },
    {
      "entropy": 5.179564189910889,
      "epoch": 1.1501837621115938,
      "grad_norm": 1.25,
      "learning_rate": 0.0004872772132923021,
      "loss": 4.9535,
      "mean_token_accuracy": 0.1964317664504051,
      "num_tokens": 25332707.0,
      "step": 13770
    },
    {
      "entropy": 5.161172771453858,
      "epoch": 1.1506014032743068,
      "grad_norm": 1.453125,
      "learning_rate": 0.00048726734260725186,
      "loss": 4.969,
      "mean_token_accuracy": 0.20147607028484343,
      "num_tokens": 25342029.0,
      "step": 13775
    },
    {
      "entropy": 5.143619251251221,
      "epoch": 1.1510190444370196,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004872574682062063,
      "loss": 4.8953,
      "mean_token_accuracy": 0.21496028453111649,
      "num_tokens": 25350374.0,
      "step": 13780
    },
    {
      "entropy": 5.203275156021118,
      "epoch": 1.1514366855997327,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004872475900893382,
      "loss": 5.0439,
      "mean_token_accuracy": 0.1992451563477516,
      "num_tokens": 25359370.0,
      "step": 13785
    },
    {
      "entropy": 5.186658763885498,
      "epoch": 1.1518543267624457,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004872377082568206,
      "loss": 4.9681,
      "mean_token_accuracy": 0.20525214672088624,
      "num_tokens": 25368329.0,
      "step": 13790
    },
    {
      "entropy": 5.210053110122681,
      "epoch": 1.1522719679251587,
      "grad_norm": 1.515625,
      "learning_rate": 0.00048722782270882643,
      "loss": 5.0835,
      "mean_token_accuracy": 0.19691489934921264,
      "num_tokens": 25377795.0,
      "step": 13795
    },
    {
      "entropy": 5.218312072753906,
      "epoch": 1.1526896090878718,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004872179334455288,
      "loss": 5.0418,
      "mean_token_accuracy": 0.20330708026885985,
      "num_tokens": 25387134.0,
      "step": 13800
    },
    {
      "entropy": 5.262246751785279,
      "epoch": 1.1531072502505846,
      "grad_norm": 1.28125,
      "learning_rate": 0.00048720804046710074,
      "loss": 5.0734,
      "mean_token_accuracy": 0.19735973775386811,
      "num_tokens": 25396320.0,
      "step": 13805
    },
    {
      "entropy": 5.262807607650757,
      "epoch": 1.1535248914132976,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00048719814377371553,
      "loss": 5.0284,
      "mean_token_accuracy": 0.19932361245155333,
      "num_tokens": 25404727.0,
      "step": 13810
    },
    {
      "entropy": 5.143845558166504,
      "epoch": 1.1539425325760106,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004871882433655464,
      "loss": 4.9351,
      "mean_token_accuracy": 0.20064099431037902,
      "num_tokens": 25414137.0,
      "step": 13815
    },
    {
      "entropy": 5.2173237800598145,
      "epoch": 1.1543601737387237,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004871783392427666,
      "loss": 5.0162,
      "mean_token_accuracy": 0.2022243246436119,
      "num_tokens": 25423297.0,
      "step": 13820
    },
    {
      "entropy": 5.320109176635742,
      "epoch": 1.1547778149014367,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004871684314055497,
      "loss": 5.0931,
      "mean_token_accuracy": 0.19940237700939178,
      "num_tokens": 25433185.0,
      "step": 13825
    },
    {
      "entropy": 5.2336243152618405,
      "epoch": 1.1551954560641498,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004871585198540689,
      "loss": 5.0181,
      "mean_token_accuracy": 0.20757358372211457,
      "num_tokens": 25441769.0,
      "step": 13830
    },
    {
      "entropy": 5.14635214805603,
      "epoch": 1.1556130972268628,
      "grad_norm": 1.34375,
      "learning_rate": 0.00048714860458849794,
      "loss": 4.9492,
      "mean_token_accuracy": 0.19906765967607498,
      "num_tokens": 25450650.0,
      "step": 13835
    },
    {
      "entropy": 5.148273849487305,
      "epoch": 1.1560307383895756,
      "grad_norm": 1.296875,
      "learning_rate": 0.00048713868560901013,
      "loss": 4.9501,
      "mean_token_accuracy": 0.21477721780538558,
      "num_tokens": 25458721.0,
      "step": 13840
    },
    {
      "entropy": 5.216542673110962,
      "epoch": 1.1564483795522886,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004871287629157794,
      "loss": 5.0193,
      "mean_token_accuracy": 0.20267477482557297,
      "num_tokens": 25468004.0,
      "step": 13845
    },
    {
      "entropy": 5.246415615081787,
      "epoch": 1.1568660207150017,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004871188365089793,
      "loss": 5.1334,
      "mean_token_accuracy": 0.19461865425109864,
      "num_tokens": 25478042.0,
      "step": 13850
    },
    {
      "entropy": 5.281588220596314,
      "epoch": 1.1572836618777147,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004871089063887837,
      "loss": 5.0399,
      "mean_token_accuracy": 0.19607396125793458,
      "num_tokens": 25488439.0,
      "step": 13855
    },
    {
      "entropy": 5.200211381912231,
      "epoch": 1.1577013030404277,
      "grad_norm": 1.421875,
      "learning_rate": 0.00048709897255536626,
      "loss": 5.0364,
      "mean_token_accuracy": 0.19912295937538146,
      "num_tokens": 25497928.0,
      "step": 13860
    },
    {
      "entropy": 5.2542280673980715,
      "epoch": 1.1581189442031405,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004870890350089011,
      "loss": 5.0475,
      "mean_token_accuracy": 0.20107868164777756,
      "num_tokens": 25506522.0,
      "step": 13865
    },
    {
      "entropy": 5.267274808883667,
      "epoch": 1.1585365853658536,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004870790937495621,
      "loss": 5.0453,
      "mean_token_accuracy": 0.18750682175159455,
      "num_tokens": 25516017.0,
      "step": 13870
    },
    {
      "entropy": 5.262888193130493,
      "epoch": 1.1589542265285666,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00048706914877752325,
      "loss": 5.0,
      "mean_token_accuracy": 0.2010779470205307,
      "num_tokens": 25525262.0,
      "step": 13875
    },
    {
      "entropy": 5.11389799118042,
      "epoch": 1.1593718676912796,
      "grad_norm": 1.3125,
      "learning_rate": 0.00048705920009295867,
      "loss": 4.9868,
      "mean_token_accuracy": 0.20504505783319474,
      "num_tokens": 25534479.0,
      "step": 13880
    },
    {
      "entropy": 5.1637506008148195,
      "epoch": 1.1597895088539927,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004870492476960426,
      "loss": 5.0109,
      "mean_token_accuracy": 0.2007748156785965,
      "num_tokens": 25543338.0,
      "step": 13885
    },
    {
      "entropy": 5.334021472930909,
      "epoch": 1.1602071500167057,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004870392915869491,
      "loss": 5.1263,
      "mean_token_accuracy": 0.19538510739803314,
      "num_tokens": 25552286.0,
      "step": 13890
    },
    {
      "entropy": 5.258430051803589,
      "epoch": 1.1606247911794187,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004870293317658526,
      "loss": 5.0507,
      "mean_token_accuracy": 0.1918945714831352,
      "num_tokens": 25561978.0,
      "step": 13895
    },
    {
      "entropy": 5.156156158447265,
      "epoch": 1.1610424323421316,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004870193682329275,
      "loss": 4.9487,
      "mean_token_accuracy": 0.1999768301844597,
      "num_tokens": 25570139.0,
      "step": 13900
    },
    {
      "entropy": 5.2067883014678955,
      "epoch": 1.1614600735048446,
      "grad_norm": 1.625,
      "learning_rate": 0.00048700940098834805,
      "loss": 4.9855,
      "mean_token_accuracy": 0.19982839971780778,
      "num_tokens": 25579761.0,
      "step": 13905
    },
    {
      "entropy": 5.158232116699219,
      "epoch": 1.1618777146675576,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004869994300322889,
      "loss": 4.993,
      "mean_token_accuracy": 0.20249005705118178,
      "num_tokens": 25588740.0,
      "step": 13910
    },
    {
      "entropy": 5.1750349521636965,
      "epoch": 1.1622953558302707,
      "grad_norm": 1.328125,
      "learning_rate": 0.00048698945536492454,
      "loss": 5.017,
      "mean_token_accuracy": 0.19673240929841995,
      "num_tokens": 25597963.0,
      "step": 13915
    },
    {
      "entropy": 5.184742975234985,
      "epoch": 1.1627129969929837,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004869794769864296,
      "loss": 4.9807,
      "mean_token_accuracy": 0.20275705456733703,
      "num_tokens": 25606782.0,
      "step": 13920
    },
    {
      "entropy": 5.303674173355103,
      "epoch": 1.1631306381556965,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004869694948969787,
      "loss": 5.1061,
      "mean_token_accuracy": 0.19323970824480058,
      "num_tokens": 25616458.0,
      "step": 13925
    },
    {
      "entropy": 5.26280632019043,
      "epoch": 1.1635482793184095,
      "grad_norm": 1.25,
      "learning_rate": 0.00048695950909674666,
      "loss": 5.0474,
      "mean_token_accuracy": 0.1922272726893425,
      "num_tokens": 25625081.0,
      "step": 13930
    },
    {
      "entropy": 5.245468616485596,
      "epoch": 1.1639659204811226,
      "grad_norm": 7.5,
      "learning_rate": 0.00048694951958590835,
      "loss": 5.0727,
      "mean_token_accuracy": 0.19748628735542298,
      "num_tokens": 25634000.0,
      "step": 13935
    },
    {
      "entropy": 5.240093517303467,
      "epoch": 1.1643835616438356,
      "grad_norm": 1.484375,
      "learning_rate": 0.00048693952636463854,
      "loss": 4.9548,
      "mean_token_accuracy": 0.2037039428949356,
      "num_tokens": 25642715.0,
      "step": 13940
    },
    {
      "entropy": 5.085875844955444,
      "epoch": 1.1648012028065486,
      "grad_norm": 1.40625,
      "learning_rate": 0.00048692952943311214,
      "loss": 4.7905,
      "mean_token_accuracy": 0.22164323031902314,
      "num_tokens": 25650979.0,
      "step": 13945
    },
    {
      "entropy": 5.178522348403931,
      "epoch": 1.1652188439692617,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004869195287915043,
      "loss": 5.0362,
      "mean_token_accuracy": 0.19704795628786087,
      "num_tokens": 25660860.0,
      "step": 13950
    },
    {
      "entropy": 5.0678017139434814,
      "epoch": 1.1656364851319747,
      "grad_norm": 1.7734375,
      "learning_rate": 0.00048690952443999,
      "loss": 4.8369,
      "mean_token_accuracy": 0.20548441261053085,
      "num_tokens": 25670331.0,
      "step": 13955
    },
    {
      "entropy": 5.27651834487915,
      "epoch": 1.1660541262946875,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00048689951637874445,
      "loss": 5.1068,
      "mean_token_accuracy": 0.19364058524370192,
      "num_tokens": 25679276.0,
      "step": 13960
    },
    {
      "entropy": 5.244553184509277,
      "epoch": 1.1664717674574006,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004868895046079428,
      "loss": 5.0195,
      "mean_token_accuracy": 0.20563008338212968,
      "num_tokens": 25688509.0,
      "step": 13965
    },
    {
      "entropy": 5.094649696350098,
      "epoch": 1.1668894086201136,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004868794891277602,
      "loss": 4.8652,
      "mean_token_accuracy": 0.20790226012468338,
      "num_tokens": 25697776.0,
      "step": 13970
    },
    {
      "entropy": 5.178790521621704,
      "epoch": 1.1673070497828266,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004868694699383722,
      "loss": 5.0425,
      "mean_token_accuracy": 0.19789552837610244,
      "num_tokens": 25707528.0,
      "step": 13975
    },
    {
      "entropy": 5.316372585296631,
      "epoch": 1.1677246909455397,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004868594470399541,
      "loss": 4.9923,
      "mean_token_accuracy": 0.20014701038599014,
      "num_tokens": 25716568.0,
      "step": 13980
    },
    {
      "entropy": 5.171524715423584,
      "epoch": 1.1681423321082525,
      "grad_norm": 1.359375,
      "learning_rate": 0.00048684942043268137,
      "loss": 5.0271,
      "mean_token_accuracy": 0.19953645914793014,
      "num_tokens": 25726441.0,
      "step": 13985
    },
    {
      "entropy": 5.087814474105835,
      "epoch": 1.1685599732709655,
      "grad_norm": 1.328125,
      "learning_rate": 0.00048683939011672953,
      "loss": 5.0031,
      "mean_token_accuracy": 0.20528616905212402,
      "num_tokens": 25735386.0,
      "step": 13990
    },
    {
      "entropy": 5.2520115852355955,
      "epoch": 1.1689776144336785,
      "grad_norm": 1.7265625,
      "learning_rate": 0.00048682935609227414,
      "loss": 4.9641,
      "mean_token_accuracy": 0.1966576486825943,
      "num_tokens": 25744758.0,
      "step": 13995
    },
    {
      "entropy": 5.158217668533325,
      "epoch": 1.1693952555963916,
      "grad_norm": 1.484375,
      "learning_rate": 0.00048681931835949083,
      "loss": 5.0254,
      "mean_token_accuracy": 0.20261257737874985,
      "num_tokens": 25753836.0,
      "step": 14000
    },
    {
      "entropy": 5.214709949493408,
      "epoch": 1.1698128967591046,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004868092769185555,
      "loss": 5.043,
      "mean_token_accuracy": 0.18973105549812316,
      "num_tokens": 25763302.0,
      "step": 14005
    },
    {
      "entropy": 5.2364709854125975,
      "epoch": 1.1702305379218176,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00048679923176964384,
      "loss": 4.9214,
      "mean_token_accuracy": 0.2069272056221962,
      "num_tokens": 25772204.0,
      "step": 14010
    },
    {
      "entropy": 5.201098442077637,
      "epoch": 1.1706481790845307,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004867891829129316,
      "loss": 4.9934,
      "mean_token_accuracy": 0.19857273399829864,
      "num_tokens": 25781762.0,
      "step": 14015
    },
    {
      "entropy": 5.1591877937316895,
      "epoch": 1.1710658202472435,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004867791303485948,
      "loss": 5.0581,
      "mean_token_accuracy": 0.19061051607131957,
      "num_tokens": 25791573.0,
      "step": 14020
    },
    {
      "entropy": 5.200064182281494,
      "epoch": 1.1714834614099565,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004867690740768094,
      "loss": 5.0064,
      "mean_token_accuracy": 0.2024068772792816,
      "num_tokens": 25800777.0,
      "step": 14025
    },
    {
      "entropy": 5.205181455612182,
      "epoch": 1.1719011025726696,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00048675901409775147,
      "loss": 4.9712,
      "mean_token_accuracy": 0.19462422728538514,
      "num_tokens": 25809145.0,
      "step": 14030
    },
    {
      "entropy": 5.128767251968384,
      "epoch": 1.1723187437353826,
      "grad_norm": 1.25,
      "learning_rate": 0.00048674895041159714,
      "loss": 4.8892,
      "mean_token_accuracy": 0.20848254561424256,
      "num_tokens": 25817884.0,
      "step": 14035
    },
    {
      "entropy": 5.2021763801574705,
      "epoch": 1.1727363848980956,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00048673888301852247,
      "loss": 5.0093,
      "mean_token_accuracy": 0.20751434117555617,
      "num_tokens": 25826953.0,
      "step": 14040
    },
    {
      "entropy": 5.201744651794433,
      "epoch": 1.1731540260608084,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004867288119187039,
      "loss": 5.0417,
      "mean_token_accuracy": 0.1993072047829628,
      "num_tokens": 25835597.0,
      "step": 14045
    },
    {
      "entropy": 5.19501428604126,
      "epoch": 1.1735716672235215,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004867187371123175,
      "loss": 5.0312,
      "mean_token_accuracy": 0.1943700760602951,
      "num_tokens": 25845072.0,
      "step": 14050
    },
    {
      "entropy": 5.213809776306152,
      "epoch": 1.1739893083862345,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00048670865859953984,
      "loss": 4.9134,
      "mean_token_accuracy": 0.2005431592464447,
      "num_tokens": 25854902.0,
      "step": 14055
    },
    {
      "entropy": 5.187883377075195,
      "epoch": 1.1744069495489475,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004866985763805472,
      "loss": 5.0628,
      "mean_token_accuracy": 0.19347578287124634,
      "num_tokens": 25863807.0,
      "step": 14060
    },
    {
      "entropy": 5.239105129241944,
      "epoch": 1.1748245907116606,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004866884904555162,
      "loss": 5.048,
      "mean_token_accuracy": 0.19782054126262666,
      "num_tokens": 25873709.0,
      "step": 14065
    },
    {
      "entropy": 5.196970510482788,
      "epoch": 1.1752422318743736,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004866784008246233,
      "loss": 5.0158,
      "mean_token_accuracy": 0.19895295202732086,
      "num_tokens": 25883425.0,
      "step": 14070
    },
    {
      "entropy": 5.1306873798370365,
      "epoch": 1.1756598730370866,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004866683074880453,
      "loss": 4.903,
      "mean_token_accuracy": 0.20663238316774368,
      "num_tokens": 25892317.0,
      "step": 14075
    },
    {
      "entropy": 5.149318981170654,
      "epoch": 1.1760775141997994,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00048665821044595884,
      "loss": 4.9655,
      "mean_token_accuracy": 0.2042432978749275,
      "num_tokens": 25901733.0,
      "step": 14080
    },
    {
      "entropy": 5.111167049407959,
      "epoch": 1.1764951553625125,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004866481096985405,
      "loss": 4.9645,
      "mean_token_accuracy": 0.20249925106763839,
      "num_tokens": 25911574.0,
      "step": 14085
    },
    {
      "entropy": 5.192024517059326,
      "epoch": 1.1769127965252255,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004866380052459673,
      "loss": 5.0416,
      "mean_token_accuracy": 0.19624707847833633,
      "num_tokens": 25921152.0,
      "step": 14090
    },
    {
      "entropy": 5.317298746109008,
      "epoch": 1.1773304376879385,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004866278970884161,
      "loss": 5.0053,
      "mean_token_accuracy": 0.19719336032867432,
      "num_tokens": 25930455.0,
      "step": 14095
    },
    {
      "entropy": 5.194132709503174,
      "epoch": 1.1777480788506516,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00048661778522606376,
      "loss": 4.9229,
      "mean_token_accuracy": 0.20312894880771637,
      "num_tokens": 25939298.0,
      "step": 14100
    },
    {
      "entropy": 5.12615213394165,
      "epoch": 1.1781657200133644,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00048660766965908736,
      "loss": 4.9941,
      "mean_token_accuracy": 0.19395186752080917,
      "num_tokens": 25948426.0,
      "step": 14105
    },
    {
      "entropy": 5.243398141860962,
      "epoch": 1.1785833611760774,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00048659755038766406,
      "loss": 4.9907,
      "mean_token_accuracy": 0.20139980763196946,
      "num_tokens": 25958469.0,
      "step": 14110
    },
    {
      "entropy": 5.156703090667724,
      "epoch": 1.1790010023387905,
      "grad_norm": 1.28125,
      "learning_rate": 0.00048658742741197086,
      "loss": 4.8811,
      "mean_token_accuracy": 0.21628623455762863,
      "num_tokens": 25967173.0,
      "step": 14115
    },
    {
      "entropy": 5.16376953125,
      "epoch": 1.1794186435015035,
      "grad_norm": 1.296875,
      "learning_rate": 0.00048657730073218517,
      "loss": 5.023,
      "mean_token_accuracy": 0.19851044267416001,
      "num_tokens": 25976474.0,
      "step": 14120
    },
    {
      "entropy": 5.190597200393677,
      "epoch": 1.1798362846642165,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004865671703484841,
      "loss": 5.0046,
      "mean_token_accuracy": 0.20353392213582994,
      "num_tokens": 25984745.0,
      "step": 14125
    },
    {
      "entropy": 5.184911680221558,
      "epoch": 1.1802539258269296,
      "grad_norm": 1.203125,
      "learning_rate": 0.00048655703626104493,
      "loss": 4.9158,
      "mean_token_accuracy": 0.20863913297653197,
      "num_tokens": 25993265.0,
      "step": 14130
    },
    {
      "entropy": 5.1855055809021,
      "epoch": 1.1806715669896426,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00048654689847004535,
      "loss": 4.9533,
      "mean_token_accuracy": 0.20211062729358673,
      "num_tokens": 26002390.0,
      "step": 14135
    },
    {
      "entropy": 5.279131841659546,
      "epoch": 1.1810892081523554,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004865367569756626,
      "loss": 5.1252,
      "mean_token_accuracy": 0.19882426410913467,
      "num_tokens": 26013086.0,
      "step": 14140
    },
    {
      "entropy": 5.278318738937378,
      "epoch": 1.1815068493150684,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00048652661177807426,
      "loss": 5.0069,
      "mean_token_accuracy": 0.20200712531805037,
      "num_tokens": 26022119.0,
      "step": 14145
    },
    {
      "entropy": 5.228790426254273,
      "epoch": 1.1819244904777815,
      "grad_norm": 1.328125,
      "learning_rate": 0.00048651646287745803,
      "loss": 5.0533,
      "mean_token_accuracy": 0.19350807517766952,
      "num_tokens": 26031805.0,
      "step": 14150
    },
    {
      "entropy": 5.1865636825561525,
      "epoch": 1.1823421316404945,
      "grad_norm": 1.21875,
      "learning_rate": 0.00048650631027399155,
      "loss": 5.0263,
      "mean_token_accuracy": 0.19875009208917618,
      "num_tokens": 26040965.0,
      "step": 14155
    },
    {
      "entropy": 5.202555751800537,
      "epoch": 1.1827597728032075,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00048649615396785246,
      "loss": 5.0011,
      "mean_token_accuracy": 0.2021612271666527,
      "num_tokens": 26049618.0,
      "step": 14160
    },
    {
      "entropy": 5.181373167037964,
      "epoch": 1.1831774139659206,
      "grad_norm": 1.171875,
      "learning_rate": 0.00048648599395921863,
      "loss": 4.9447,
      "mean_token_accuracy": 0.20698685795068741,
      "num_tokens": 26058335.0,
      "step": 14165
    },
    {
      "entropy": 5.068678379058838,
      "epoch": 1.1835950551286334,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00048647583024826786,
      "loss": 4.8397,
      "mean_token_accuracy": 0.21249986439943314,
      "num_tokens": 26067361.0,
      "step": 14170
    },
    {
      "entropy": 5.249664926528931,
      "epoch": 1.1840126962913464,
      "grad_norm": 1.3125,
      "learning_rate": 0.00048646566283517824,
      "loss": 5.0746,
      "mean_token_accuracy": 0.19270857572555541,
      "num_tokens": 26077401.0,
      "step": 14175
    },
    {
      "entropy": 5.13611478805542,
      "epoch": 1.1844303374540595,
      "grad_norm": 1.359375,
      "learning_rate": 0.00048645549172012756,
      "loss": 4.902,
      "mean_token_accuracy": 0.20992079228162766,
      "num_tokens": 26086422.0,
      "step": 14180
    },
    {
      "entropy": 5.160061264038086,
      "epoch": 1.1848479786167725,
      "grad_norm": 1.203125,
      "learning_rate": 0.000486445316903294,
      "loss": 5.051,
      "mean_token_accuracy": 0.1988341450691223,
      "num_tokens": 26095974.0,
      "step": 14185
    },
    {
      "entropy": 5.167821741104126,
      "epoch": 1.1852656197794855,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004864351383848556,
      "loss": 4.9462,
      "mean_token_accuracy": 0.20239839553833008,
      "num_tokens": 26104782.0,
      "step": 14190
    },
    {
      "entropy": 5.266237831115722,
      "epoch": 1.1856832609421986,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004864249561649906,
      "loss": 5.0107,
      "mean_token_accuracy": 0.20159432291984558,
      "num_tokens": 26113769.0,
      "step": 14195
    },
    {
      "entropy": 5.239569759368896,
      "epoch": 1.1861009021049114,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004864147702438773,
      "loss": 5.0401,
      "mean_token_accuracy": 0.19899984300136567,
      "num_tokens": 26123759.0,
      "step": 14200
    },
    {
      "entropy": 5.180622291564942,
      "epoch": 1.1865185432676244,
      "grad_norm": 1.203125,
      "learning_rate": 0.000486404580621694,
      "loss": 4.9264,
      "mean_token_accuracy": 0.21200331449508666,
      "num_tokens": 26133392.0,
      "step": 14205
    },
    {
      "entropy": 5.207076740264893,
      "epoch": 1.1869361844303374,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000486394387298619,
      "loss": 5.0927,
      "mean_token_accuracy": 0.19528756588697432,
      "num_tokens": 26142102.0,
      "step": 14210
    },
    {
      "entropy": 5.153818559646607,
      "epoch": 1.1873538255930505,
      "grad_norm": 1.296875,
      "learning_rate": 0.00048638419027483075,
      "loss": 4.9663,
      "mean_token_accuracy": 0.20160269290208815,
      "num_tokens": 26151504.0,
      "step": 14215
    },
    {
      "entropy": 5.194489526748657,
      "epoch": 1.1877714667557635,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004863739895505079,
      "loss": 5.1079,
      "mean_token_accuracy": 0.192075414955616,
      "num_tokens": 26160717.0,
      "step": 14220
    },
    {
      "entropy": 5.223006916046143,
      "epoch": 1.1881891079184765,
      "grad_norm": 1.359375,
      "learning_rate": 0.00048636378512582884,
      "loss": 4.9532,
      "mean_token_accuracy": 0.2052110478281975,
      "num_tokens": 26169788.0,
      "step": 14225
    },
    {
      "entropy": 5.187114810943603,
      "epoch": 1.1886067490811894,
      "grad_norm": 1.296875,
      "learning_rate": 0.00048635357700097236,
      "loss": 4.9381,
      "mean_token_accuracy": 0.19482695013284684,
      "num_tokens": 26178904.0,
      "step": 14230
    },
    {
      "entropy": 5.225716161727905,
      "epoch": 1.1890243902439024,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004863433651761171,
      "loss": 5.0225,
      "mean_token_accuracy": 0.19539916664361953,
      "num_tokens": 26188832.0,
      "step": 14235
    },
    {
      "entropy": 5.212666368484497,
      "epoch": 1.1894420314066154,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004863331496514418,
      "loss": 5.1005,
      "mean_token_accuracy": 0.19729874581098555,
      "num_tokens": 26199003.0,
      "step": 14240
    },
    {
      "entropy": 5.27499418258667,
      "epoch": 1.1898596725693285,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048632293042712545,
      "loss": 4.9442,
      "mean_token_accuracy": 0.200897416472435,
      "num_tokens": 26209001.0,
      "step": 14245
    },
    {
      "entropy": 5.223832702636718,
      "epoch": 1.1902773137320415,
      "grad_norm": 1.3125,
      "learning_rate": 0.00048631270750334674,
      "loss": 4.9999,
      "mean_token_accuracy": 0.20624465644359588,
      "num_tokens": 26217834.0,
      "step": 14250
    },
    {
      "entropy": 5.2098418235778805,
      "epoch": 1.1906949548947545,
      "grad_norm": 1.421875,
      "learning_rate": 0.00048630248088028474,
      "loss": 5.0408,
      "mean_token_accuracy": 0.19911759495735168,
      "num_tokens": 26226935.0,
      "step": 14255
    },
    {
      "entropy": 5.218213939666748,
      "epoch": 1.1911125960574673,
      "grad_norm": 1.25,
      "learning_rate": 0.0004862922505581185,
      "loss": 5.0347,
      "mean_token_accuracy": 0.19952414482831954,
      "num_tokens": 26236874.0,
      "step": 14260
    },
    {
      "entropy": 5.233570194244384,
      "epoch": 1.1915302372201804,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000486282016537027,
      "loss": 4.9808,
      "mean_token_accuracy": 0.2077031835913658,
      "num_tokens": 26246361.0,
      "step": 14265
    },
    {
      "entropy": 5.127079820632934,
      "epoch": 1.1919478783828934,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00048627177881718954,
      "loss": 4.924,
      "mean_token_accuracy": 0.2085671380162239,
      "num_tokens": 26255316.0,
      "step": 14270
    },
    {
      "entropy": 5.314019346237183,
      "epoch": 1.1923655195456064,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004862615373987853,
      "loss": 5.0647,
      "mean_token_accuracy": 0.20198653191328048,
      "num_tokens": 26265250.0,
      "step": 14275
    },
    {
      "entropy": 5.115083265304565,
      "epoch": 1.1927831607083195,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004862512922819935,
      "loss": 4.8445,
      "mean_token_accuracy": 0.2087085574865341,
      "num_tokens": 26273858.0,
      "step": 14280
    },
    {
      "entropy": 5.198866844177246,
      "epoch": 1.1932008018710325,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004862410434669936,
      "loss": 5.0857,
      "mean_token_accuracy": 0.1963450714945793,
      "num_tokens": 26283536.0,
      "step": 14285
    },
    {
      "entropy": 5.181536245346069,
      "epoch": 1.1936184430337453,
      "grad_norm": 1.3125,
      "learning_rate": 0.00048623079095396496,
      "loss": 4.9168,
      "mean_token_accuracy": 0.21034623086452484,
      "num_tokens": 26291761.0,
      "step": 14290
    },
    {
      "entropy": 5.172656011581421,
      "epoch": 1.1940360841964583,
      "grad_norm": 1.6640625,
      "learning_rate": 0.000486220534743087,
      "loss": 4.9544,
      "mean_token_accuracy": 0.19319161921739578,
      "num_tokens": 26301082.0,
      "step": 14295
    },
    {
      "entropy": 5.12870831489563,
      "epoch": 1.1944537253591714,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004862102748345394,
      "loss": 5.0317,
      "mean_token_accuracy": 0.19517704099416733,
      "num_tokens": 26310424.0,
      "step": 14300
    },
    {
      "entropy": 5.246632528305054,
      "epoch": 1.1948713665218844,
      "grad_norm": 1.3125,
      "learning_rate": 0.00048620001122850164,
      "loss": 5.0918,
      "mean_token_accuracy": 0.19533061981201172,
      "num_tokens": 26319442.0,
      "step": 14305
    },
    {
      "entropy": 5.164397430419922,
      "epoch": 1.1952890076845974,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004861897439251535,
      "loss": 4.9159,
      "mean_token_accuracy": 0.20808929204940796,
      "num_tokens": 26327757.0,
      "step": 14310
    },
    {
      "entropy": 5.160078859329223,
      "epoch": 1.1957066488473105,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00048617947292467473,
      "loss": 4.9633,
      "mean_token_accuracy": 0.19601308256387712,
      "num_tokens": 26337327.0,
      "step": 14315
    },
    {
      "entropy": 5.180516147613526,
      "epoch": 1.1961242900100233,
      "grad_norm": 1.3125,
      "learning_rate": 0.00048616919822724506,
      "loss": 5.06,
      "mean_token_accuracy": 0.2019834563136101,
      "num_tokens": 26345821.0,
      "step": 14320
    },
    {
      "entropy": 5.2360007762908936,
      "epoch": 1.1965419311727363,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00048615891983304436,
      "loss": 5.0091,
      "mean_token_accuracy": 0.1995691552758217,
      "num_tokens": 26355423.0,
      "step": 14325
    },
    {
      "entropy": 5.14224910736084,
      "epoch": 1.1969595723354494,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004861486377422526,
      "loss": 4.8823,
      "mean_token_accuracy": 0.21077274680137634,
      "num_tokens": 26363770.0,
      "step": 14330
    },
    {
      "entropy": 5.109952449798584,
      "epoch": 1.1973772134981624,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004861383519550498,
      "loss": 4.9358,
      "mean_token_accuracy": 0.20614567548036575,
      "num_tokens": 26373530.0,
      "step": 14335
    },
    {
      "entropy": 5.303129529953003,
      "epoch": 1.1977948546608754,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00048612806247161594,
      "loss": 5.1175,
      "mean_token_accuracy": 0.18702830523252487,
      "num_tokens": 26382960.0,
      "step": 14340
    },
    {
      "entropy": 5.176595401763916,
      "epoch": 1.1982124958235885,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004861177692921313,
      "loss": 4.8902,
      "mean_token_accuracy": 0.21062125116586686,
      "num_tokens": 26392105.0,
      "step": 14345
    },
    {
      "entropy": 5.170688629150391,
      "epoch": 1.1986301369863013,
      "grad_norm": 1.484375,
      "learning_rate": 0.00048610747241677595,
      "loss": 5.006,
      "mean_token_accuracy": 0.20741214901208876,
      "num_tokens": 26401780.0,
      "step": 14350
    },
    {
      "entropy": 5.203456544876099,
      "epoch": 1.1990477781490143,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004860971718457302,
      "loss": 4.9884,
      "mean_token_accuracy": 0.2039581447839737,
      "num_tokens": 26410290.0,
      "step": 14355
    },
    {
      "entropy": 5.2359918594360355,
      "epoch": 1.1994654193117273,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004860868675791743,
      "loss": 5.0132,
      "mean_token_accuracy": 0.2057497203350067,
      "num_tokens": 26420591.0,
      "step": 14360
    },
    {
      "entropy": 5.217155265808105,
      "epoch": 1.1998830604744404,
      "grad_norm": 1.375,
      "learning_rate": 0.00048607655961728876,
      "loss": 5.0132,
      "mean_token_accuracy": 0.19439144134521485,
      "num_tokens": 26429453.0,
      "step": 14365
    },
    {
      "entropy": 5.258237266540528,
      "epoch": 1.2003007016371534,
      "grad_norm": 1.484375,
      "learning_rate": 0.000486066247960254,
      "loss": 5.0156,
      "mean_token_accuracy": 0.201152004301548,
      "num_tokens": 26439111.0,
      "step": 14370
    },
    {
      "entropy": 5.147270011901855,
      "epoch": 1.2007183427998664,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048605593260825044,
      "loss": 4.9351,
      "mean_token_accuracy": 0.20845318585634232,
      "num_tokens": 26448071.0,
      "step": 14375
    },
    {
      "entropy": 5.215981435775757,
      "epoch": 1.2011359839625793,
      "grad_norm": 1.421875,
      "learning_rate": 0.00048604561356145875,
      "loss": 5.1094,
      "mean_token_accuracy": 0.19657070338726043,
      "num_tokens": 26457226.0,
      "step": 14380
    },
    {
      "entropy": 5.232178783416748,
      "epoch": 1.2015536251252923,
      "grad_norm": 1.625,
      "learning_rate": 0.00048603529082005955,
      "loss": 4.9968,
      "mean_token_accuracy": 0.20379756540060043,
      "num_tokens": 26466052.0,
      "step": 14385
    },
    {
      "entropy": 5.249006128311157,
      "epoch": 1.2019712662880053,
      "grad_norm": 1.515625,
      "learning_rate": 0.00048602496438423354,
      "loss": 4.9943,
      "mean_token_accuracy": 0.1991071954369545,
      "num_tokens": 26476102.0,
      "step": 14390
    },
    {
      "entropy": 5.190280294418335,
      "epoch": 1.2023889074507184,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004860146342541616,
      "loss": 5.0085,
      "mean_token_accuracy": 0.20177917033433915,
      "num_tokens": 26485775.0,
      "step": 14395
    },
    {
      "entropy": 5.1901891231536865,
      "epoch": 1.2028065486134314,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004860043004300244,
      "loss": 4.9624,
      "mean_token_accuracy": 0.19902112036943437,
      "num_tokens": 26495268.0,
      "step": 14400
    },
    {
      "entropy": 5.236022806167602,
      "epoch": 1.2032241897761444,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000485993962912003,
      "loss": 5.0343,
      "mean_token_accuracy": 0.2016926258802414,
      "num_tokens": 26505253.0,
      "step": 14405
    },
    {
      "entropy": 5.184625911712646,
      "epoch": 1.2036418309388572,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004859836217002783,
      "loss": 4.9919,
      "mean_token_accuracy": 0.20034309923648835,
      "num_tokens": 26513609.0,
      "step": 14410
    },
    {
      "entropy": 5.26208815574646,
      "epoch": 1.2040594721015703,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004859732767950313,
      "loss": 5.0486,
      "mean_token_accuracy": 0.1933481827378273,
      "num_tokens": 26523565.0,
      "step": 14415
    },
    {
      "entropy": 5.216074132919312,
      "epoch": 1.2044771132642833,
      "grad_norm": 1.28125,
      "learning_rate": 0.00048596292819644315,
      "loss": 4.9297,
      "mean_token_accuracy": 0.2077425554394722,
      "num_tokens": 26532554.0,
      "step": 14420
    },
    {
      "entropy": 5.181769466400146,
      "epoch": 1.2048947544269963,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000485952575904695,
      "loss": 5.0645,
      "mean_token_accuracy": 0.1949892073869705,
      "num_tokens": 26541474.0,
      "step": 14425
    },
    {
      "entropy": 5.113822412490845,
      "epoch": 1.2053123955897094,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00048594221991996814,
      "loss": 4.9342,
      "mean_token_accuracy": 0.20384753346443177,
      "num_tokens": 26551095.0,
      "step": 14430
    },
    {
      "entropy": 5.168420839309692,
      "epoch": 1.2057300367524224,
      "grad_norm": 1.328125,
      "learning_rate": 0.00048593186024244373,
      "loss": 4.9454,
      "mean_token_accuracy": 0.20573414117097855,
      "num_tokens": 26560144.0,
      "step": 14435
    },
    {
      "entropy": 5.116758108139038,
      "epoch": 1.2061476779151352,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004859214968723033,
      "loss": 4.8749,
      "mean_token_accuracy": 0.20939862132072448,
      "num_tokens": 26568983.0,
      "step": 14440
    },
    {
      "entropy": 5.1822192668914795,
      "epoch": 1.2065653190778483,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00048591112980972813,
      "loss": 4.909,
      "mean_token_accuracy": 0.20597486943006516,
      "num_tokens": 26577229.0,
      "step": 14445
    },
    {
      "entropy": 5.217160081863403,
      "epoch": 1.2069829602405613,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004859007590548997,
      "loss": 4.9878,
      "mean_token_accuracy": 0.19720837622880935,
      "num_tokens": 26585567.0,
      "step": 14450
    },
    {
      "entropy": 5.148251676559449,
      "epoch": 1.2074006014032743,
      "grad_norm": 1.28125,
      "learning_rate": 0.00048589038460799974,
      "loss": 5.0068,
      "mean_token_accuracy": 0.20020848363637925,
      "num_tokens": 26594656.0,
      "step": 14455
    },
    {
      "entropy": 5.153583240509033,
      "epoch": 1.2078182425659874,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004858800064692097,
      "loss": 5.0013,
      "mean_token_accuracy": 0.1906233862042427,
      "num_tokens": 26604002.0,
      "step": 14460
    },
    {
      "entropy": 5.224992942810059,
      "epoch": 1.2082358837287004,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00048586962463871124,
      "loss": 4.984,
      "mean_token_accuracy": 0.19845034480094909,
      "num_tokens": 26613376.0,
      "step": 14465
    },
    {
      "entropy": 5.299638414382935,
      "epoch": 1.2086535248914134,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00048585923911668625,
      "loss": 5.1198,
      "mean_token_accuracy": 0.19347414672374724,
      "num_tokens": 26623117.0,
      "step": 14470
    },
    {
      "entropy": 5.2145411491394045,
      "epoch": 1.2090711660541262,
      "grad_norm": 1.203125,
      "learning_rate": 0.00048584884990331644,
      "loss": 5.0419,
      "mean_token_accuracy": 0.19637006670236587,
      "num_tokens": 26632855.0,
      "step": 14475
    },
    {
      "entropy": 5.1195268630981445,
      "epoch": 1.2094888072168393,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00048583845699878375,
      "loss": 4.7934,
      "mean_token_accuracy": 0.21820642352104186,
      "num_tokens": 26641753.0,
      "step": 14480
    },
    {
      "entropy": 5.086602592468262,
      "epoch": 1.2099064483795523,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004858280604032701,
      "loss": 4.9323,
      "mean_token_accuracy": 0.20544191151857377,
      "num_tokens": 26650278.0,
      "step": 14485
    },
    {
      "entropy": 5.183471012115478,
      "epoch": 1.2103240895422653,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048581766011695745,
      "loss": 5.0105,
      "mean_token_accuracy": 0.20775025337934494,
      "num_tokens": 26658891.0,
      "step": 14490
    },
    {
      "entropy": 5.225825214385987,
      "epoch": 1.2107417307049784,
      "grad_norm": 1.25,
      "learning_rate": 0.00048580725614002785,
      "loss": 5.0392,
      "mean_token_accuracy": 0.1940358340740204,
      "num_tokens": 26668197.0,
      "step": 14495
    },
    {
      "entropy": 5.231700372695923,
      "epoch": 1.2111593718676912,
      "grad_norm": 1.28125,
      "learning_rate": 0.00048579684847266353,
      "loss": 5.0655,
      "mean_token_accuracy": 0.1964749351143837,
      "num_tokens": 26676796.0,
      "step": 14500
    },
    {
      "entropy": 5.146570158004761,
      "epoch": 1.2115770130304042,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004857864371150467,
      "loss": 4.9558,
      "mean_token_accuracy": 0.20514973551034926,
      "num_tokens": 26686237.0,
      "step": 14505
    },
    {
      "entropy": 5.169667911529541,
      "epoch": 1.2119946541931172,
      "grad_norm": 1.4375,
      "learning_rate": 0.00048577602206735943,
      "loss": 4.9463,
      "mean_token_accuracy": 0.202629916369915,
      "num_tokens": 26695931.0,
      "step": 14510
    },
    {
      "entropy": 5.148312950134278,
      "epoch": 1.2124122953558303,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004857656033297843,
      "loss": 4.9093,
      "mean_token_accuracy": 0.20532681792974472,
      "num_tokens": 26704326.0,
      "step": 14515
    },
    {
      "entropy": 5.168211126327515,
      "epoch": 1.2128299365185433,
      "grad_norm": 1.328125,
      "learning_rate": 0.00048575518090250347,
      "loss": 4.85,
      "mean_token_accuracy": 0.2090817376971245,
      "num_tokens": 26713000.0,
      "step": 14520
    },
    {
      "entropy": 5.177458095550537,
      "epoch": 1.2132475776812564,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004857447547856996,
      "loss": 4.9758,
      "mean_token_accuracy": 0.19793840050697326,
      "num_tokens": 26722014.0,
      "step": 14525
    },
    {
      "entropy": 5.180481910705566,
      "epoch": 1.2136652188439694,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004857343249795552,
      "loss": 4.9416,
      "mean_token_accuracy": 0.2022526502609253,
      "num_tokens": 26730975.0,
      "step": 14530
    },
    {
      "entropy": 5.09341344833374,
      "epoch": 1.2140828600066822,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004857238914842526,
      "loss": 5.038,
      "mean_token_accuracy": 0.19636483937501908,
      "num_tokens": 26740087.0,
      "step": 14535
    },
    {
      "entropy": 5.173520278930664,
      "epoch": 1.2145005011693952,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048571345429997473,
      "loss": 5.0082,
      "mean_token_accuracy": 0.19615557342767714,
      "num_tokens": 26749820.0,
      "step": 14540
    },
    {
      "entropy": 5.277585363388061,
      "epoch": 1.2149181423321083,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00048570301342690416,
      "loss": 5.0881,
      "mean_token_accuracy": 0.19978706687688827,
      "num_tokens": 26759850.0,
      "step": 14545
    },
    {
      "entropy": 5.122476863861084,
      "epoch": 1.2153357834948213,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004856925688652238,
      "loss": 4.8542,
      "mean_token_accuracy": 0.21461947858333588,
      "num_tokens": 26769519.0,
      "step": 14550
    },
    {
      "entropy": 5.187227582931518,
      "epoch": 1.2157534246575343,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00048568212061511634,
      "loss": 5.0186,
      "mean_token_accuracy": 0.19181180298328399,
      "num_tokens": 26779999.0,
      "step": 14555
    },
    {
      "entropy": 5.23685941696167,
      "epoch": 1.2161710658202471,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00048567166867676475,
      "loss": 4.966,
      "mean_token_accuracy": 0.1984766960144043,
      "num_tokens": 26788903.0,
      "step": 14560
    },
    {
      "entropy": 5.288030529022217,
      "epoch": 1.2165887069829602,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00048566121305035207,
      "loss": 5.076,
      "mean_token_accuracy": 0.20037996172904968,
      "num_tokens": 26799161.0,
      "step": 14565
    },
    {
      "entropy": 5.143881511688233,
      "epoch": 1.2170063481456732,
      "grad_norm": 1.203125,
      "learning_rate": 0.00048565075373606126,
      "loss": 4.9272,
      "mean_token_accuracy": 0.20638153254985808,
      "num_tokens": 26807887.0,
      "step": 14570
    },
    {
      "entropy": 5.133748626708984,
      "epoch": 1.2174239893083862,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00048564029073407545,
      "loss": 4.9109,
      "mean_token_accuracy": 0.20690165162086488,
      "num_tokens": 26816466.0,
      "step": 14575
    },
    {
      "entropy": 5.194406270980835,
      "epoch": 1.2178416304710993,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004856298240445778,
      "loss": 5.0607,
      "mean_token_accuracy": 0.1980845332145691,
      "num_tokens": 26825847.0,
      "step": 14580
    },
    {
      "entropy": 5.285407257080078,
      "epoch": 1.2182592716338123,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004856193536677515,
      "loss": 5.0304,
      "mean_token_accuracy": 0.19032817482948303,
      "num_tokens": 26835610.0,
      "step": 14585
    },
    {
      "entropy": 5.210682725906372,
      "epoch": 1.2186769127965253,
      "grad_norm": 1.203125,
      "learning_rate": 0.00048560887960377995,
      "loss": 4.9363,
      "mean_token_accuracy": 0.1982601374387741,
      "num_tokens": 26844319.0,
      "step": 14590
    },
    {
      "entropy": 5.137855577468872,
      "epoch": 1.2190945539592382,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00048559840185284647,
      "loss": 5.0425,
      "mean_token_accuracy": 0.19627588093280793,
      "num_tokens": 26853508.0,
      "step": 14595
    },
    {
      "entropy": 5.149958324432373,
      "epoch": 1.2195121951219512,
      "grad_norm": 1.28125,
      "learning_rate": 0.00048558792041513446,
      "loss": 4.9854,
      "mean_token_accuracy": 0.20349832028150558,
      "num_tokens": 26863268.0,
      "step": 14600
    },
    {
      "entropy": 5.208442354202271,
      "epoch": 1.2199298362846642,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004855774352908273,
      "loss": 4.9756,
      "mean_token_accuracy": 0.19657991528511048,
      "num_tokens": 26871879.0,
      "step": 14605
    },
    {
      "entropy": 5.212781000137329,
      "epoch": 1.2203474774473773,
      "grad_norm": 1.25,
      "learning_rate": 0.0004855669464801088,
      "loss": 5.0874,
      "mean_token_accuracy": 0.19553375542163848,
      "num_tokens": 26880920.0,
      "step": 14610
    },
    {
      "entropy": 5.2374471664428714,
      "epoch": 1.2207651186100903,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00048555645398316237,
      "loss": 5.062,
      "mean_token_accuracy": 0.1995099440217018,
      "num_tokens": 26890020.0,
      "step": 14615
    },
    {
      "entropy": 5.199272203445434,
      "epoch": 1.221182759772803,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004855459578001718,
      "loss": 5.0111,
      "mean_token_accuracy": 0.19273147881031036,
      "num_tokens": 26899486.0,
      "step": 14620
    },
    {
      "entropy": 5.127596998214722,
      "epoch": 1.2216004009355161,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004855354579313208,
      "loss": 4.999,
      "mean_token_accuracy": 0.2049153968691826,
      "num_tokens": 26908298.0,
      "step": 14625
    },
    {
      "entropy": 5.3266349792480465,
      "epoch": 1.2220180420982292,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004855249543767932,
      "loss": 5.0159,
      "mean_token_accuracy": 0.196956042945385,
      "num_tokens": 26917601.0,
      "step": 14630
    },
    {
      "entropy": 5.122921752929687,
      "epoch": 1.2224356832609422,
      "grad_norm": 1.3125,
      "learning_rate": 0.00048551444713677285,
      "loss": 4.8566,
      "mean_token_accuracy": 0.21531386375427247,
      "num_tokens": 26926368.0,
      "step": 14635
    },
    {
      "entropy": 5.11380558013916,
      "epoch": 1.2228533244236552,
      "grad_norm": 1.390625,
      "learning_rate": 0.00048550393621144374,
      "loss": 5.0305,
      "mean_token_accuracy": 0.19345635920763016,
      "num_tokens": 26935424.0,
      "step": 14640
    },
    {
      "entropy": 5.157826375961304,
      "epoch": 1.2232709655863683,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004854934216009898,
      "loss": 4.9098,
      "mean_token_accuracy": 0.2146031767129898,
      "num_tokens": 26944906.0,
      "step": 14645
    },
    {
      "entropy": 5.1121721267700195,
      "epoch": 1.2236886067490813,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004854829033055951,
      "loss": 4.8873,
      "mean_token_accuracy": 0.2112795129418373,
      "num_tokens": 26953709.0,
      "step": 14650
    },
    {
      "entropy": 5.140665578842163,
      "epoch": 1.2241062479117941,
      "grad_norm": 1.25,
      "learning_rate": 0.0004854723813254439,
      "loss": 4.9636,
      "mean_token_accuracy": 0.19989429265260697,
      "num_tokens": 26962726.0,
      "step": 14655
    },
    {
      "entropy": 5.216067790985107,
      "epoch": 1.2245238890745072,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004854618556607203,
      "loss": 5.0712,
      "mean_token_accuracy": 0.19307892471551896,
      "num_tokens": 26972519.0,
      "step": 14660
    },
    {
      "entropy": 5.187400913238525,
      "epoch": 1.2249415302372202,
      "grad_norm": 1.390625,
      "learning_rate": 0.00048545132631160856,
      "loss": 4.9329,
      "mean_token_accuracy": 0.19796740114688874,
      "num_tokens": 26981251.0,
      "step": 14665
    },
    {
      "entropy": 5.187495613098145,
      "epoch": 1.2253591713999332,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004854407932782931,
      "loss": 4.9394,
      "mean_token_accuracy": 0.20299880355596542,
      "num_tokens": 26990707.0,
      "step": 14670
    },
    {
      "entropy": 5.142419958114624,
      "epoch": 1.2257768125626463,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004854302565609582,
      "loss": 4.9319,
      "mean_token_accuracy": 0.2073827475309372,
      "num_tokens": 26999364.0,
      "step": 14675
    },
    {
      "entropy": 5.226581573486328,
      "epoch": 1.226194453725359,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00048541971615978833,
      "loss": 5.0409,
      "mean_token_accuracy": 0.19583337754011154,
      "num_tokens": 27008757.0,
      "step": 14680
    },
    {
      "entropy": 5.13408784866333,
      "epoch": 1.226612094888072,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00048540917207496806,
      "loss": 4.9272,
      "mean_token_accuracy": 0.20404252409934998,
      "num_tokens": 27017244.0,
      "step": 14685
    },
    {
      "entropy": 5.186326217651367,
      "epoch": 1.2270297360507851,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00048539862430668204,
      "loss": 5.0114,
      "mean_token_accuracy": 0.20090916901826858,
      "num_tokens": 27026527.0,
      "step": 14690
    },
    {
      "entropy": 5.158710670471192,
      "epoch": 1.2274473772134982,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048538807285511474,
      "loss": 4.932,
      "mean_token_accuracy": 0.20515777319669723,
      "num_tokens": 27034890.0,
      "step": 14695
    },
    {
      "entropy": 5.189189910888672,
      "epoch": 1.2278650183762112,
      "grad_norm": 1.3125,
      "learning_rate": 0.00048537751772045103,
      "loss": 4.9325,
      "mean_token_accuracy": 0.2087088480591774,
      "num_tokens": 27044122.0,
      "step": 14700
    },
    {
      "entropy": 5.208041667938232,
      "epoch": 1.2282826595389242,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004853669589028757,
      "loss": 4.9752,
      "mean_token_accuracy": 0.19749539643526076,
      "num_tokens": 27053929.0,
      "step": 14705
    },
    {
      "entropy": 5.175542736053467,
      "epoch": 1.2287003007016373,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004853563964025734,
      "loss": 4.947,
      "mean_token_accuracy": 0.2051151543855667,
      "num_tokens": 27062914.0,
      "step": 14710
    },
    {
      "entropy": 5.113403224945069,
      "epoch": 1.22911794186435,
      "grad_norm": 1.34375,
      "learning_rate": 0.00048534583021972926,
      "loss": 4.974,
      "mean_token_accuracy": 0.19930148124694824,
      "num_tokens": 27072121.0,
      "step": 14715
    },
    {
      "entropy": 5.112398338317871,
      "epoch": 1.2295355830270631,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004853352603545281,
      "loss": 4.9278,
      "mean_token_accuracy": 0.20363833606243134,
      "num_tokens": 27081753.0,
      "step": 14720
    },
    {
      "entropy": 5.180920743942261,
      "epoch": 1.2299532241897762,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004853246868071551,
      "loss": 4.979,
      "mean_token_accuracy": 0.19997663795948029,
      "num_tokens": 27090942.0,
      "step": 14725
    },
    {
      "entropy": 5.229444932937622,
      "epoch": 1.2303708653524892,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048531410957779533,
      "loss": 5.0227,
      "mean_token_accuracy": 0.1985974133014679,
      "num_tokens": 27100620.0,
      "step": 14730
    },
    {
      "entropy": 5.289184474945069,
      "epoch": 1.2307885065152022,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004853035286666338,
      "loss": 5.0162,
      "mean_token_accuracy": 0.20225681960582734,
      "num_tokens": 27110370.0,
      "step": 14735
    },
    {
      "entropy": 5.087170791625977,
      "epoch": 1.231206147677915,
      "grad_norm": 1.265625,
      "learning_rate": 0.000485292944073856,
      "loss": 4.9434,
      "mean_token_accuracy": 0.20882024019956588,
      "num_tokens": 27119163.0,
      "step": 14740
    },
    {
      "entropy": 5.135728454589843,
      "epoch": 1.231623788840628,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000485282355799647,
      "loss": 4.8927,
      "mean_token_accuracy": 0.20938332974910737,
      "num_tokens": 27127855.0,
      "step": 14745
    },
    {
      "entropy": 5.232671070098877,
      "epoch": 1.232041430003341,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00048527176384419217,
      "loss": 4.9773,
      "mean_token_accuracy": 0.19906123578548432,
      "num_tokens": 27136932.0,
      "step": 14750
    },
    {
      "entropy": 5.199436378479004,
      "epoch": 1.2324590711660541,
      "grad_norm": 1.3125,
      "learning_rate": 0.00048526116820767713,
      "loss": 4.9559,
      "mean_token_accuracy": 0.2053295075893402,
      "num_tokens": 27146371.0,
      "step": 14755
    },
    {
      "entropy": 5.114437675476074,
      "epoch": 1.2328767123287672,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004852505688902872,
      "loss": 4.9637,
      "mean_token_accuracy": 0.1996703028678894,
      "num_tokens": 27156473.0,
      "step": 14760
    },
    {
      "entropy": 5.258747863769531,
      "epoch": 1.2332943534914802,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004852399658922079,
      "loss": 5.0991,
      "mean_token_accuracy": 0.1950080782175064,
      "num_tokens": 27165915.0,
      "step": 14765
    },
    {
      "entropy": 5.1088793754577635,
      "epoch": 1.2337119946541932,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004852293592136251,
      "loss": 4.8902,
      "mean_token_accuracy": 0.21041838377714156,
      "num_tokens": 27174792.0,
      "step": 14770
    },
    {
      "entropy": 5.2212952136993405,
      "epoch": 1.234129635816906,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004852187488547242,
      "loss": 4.9752,
      "mean_token_accuracy": 0.20562881380319595,
      "num_tokens": 27183785.0,
      "step": 14775
    },
    {
      "entropy": 5.284130048751831,
      "epoch": 1.234547276979619,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00048520813481569116,
      "loss": 5.1249,
      "mean_token_accuracy": 0.1925940215587616,
      "num_tokens": 27193599.0,
      "step": 14780
    },
    {
      "entropy": 5.256484270095825,
      "epoch": 1.2349649181423321,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004851975170967116,
      "loss": 5.0672,
      "mean_token_accuracy": 0.1965412974357605,
      "num_tokens": 27202191.0,
      "step": 14785
    },
    {
      "entropy": 5.159860277175904,
      "epoch": 1.2353825593050451,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004851868956979715,
      "loss": 4.9906,
      "mean_token_accuracy": 0.19970670491456985,
      "num_tokens": 27210833.0,
      "step": 14790
    },
    {
      "entropy": 5.162341165542602,
      "epoch": 1.2358002004677582,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004851762706196568,
      "loss": 4.9493,
      "mean_token_accuracy": 0.19863224029541016,
      "num_tokens": 27220053.0,
      "step": 14795
    },
    {
      "entropy": 5.257050800323486,
      "epoch": 1.236217841630471,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00048516564186195353,
      "loss": 5.0682,
      "mean_token_accuracy": 0.20032899379730223,
      "num_tokens": 27229370.0,
      "step": 14800
    },
    {
      "entropy": 5.086977624893189,
      "epoch": 1.236635482793184,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00048515500942504763,
      "loss": 4.9022,
      "mean_token_accuracy": 0.2114160791039467,
      "num_tokens": 27238923.0,
      "step": 14805
    },
    {
      "entropy": 5.2325530529022215,
      "epoch": 1.237053123955897,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00048514437330912544,
      "loss": 4.9999,
      "mean_token_accuracy": 0.20099989473819732,
      "num_tokens": 27247929.0,
      "step": 14810
    },
    {
      "entropy": 5.249862813949585,
      "epoch": 1.23747076511861,
      "grad_norm": 1.375,
      "learning_rate": 0.00048513373351437294,
      "loss": 4.9702,
      "mean_token_accuracy": 0.20396916419267655,
      "num_tokens": 27257455.0,
      "step": 14815
    },
    {
      "entropy": 5.199708318710327,
      "epoch": 1.2378884062813231,
      "grad_norm": 1.3125,
      "learning_rate": 0.00048512309004097653,
      "loss": 5.0131,
      "mean_token_accuracy": 0.1940694823861122,
      "num_tokens": 27267718.0,
      "step": 14820
    },
    {
      "entropy": 5.146140670776367,
      "epoch": 1.2383060474440362,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004851124428891226,
      "loss": 4.9111,
      "mean_token_accuracy": 0.20621122419834137,
      "num_tokens": 27276706.0,
      "step": 14825
    },
    {
      "entropy": 5.075187015533447,
      "epoch": 1.2387236886067492,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004851017920589973,
      "loss": 4.8422,
      "mean_token_accuracy": 0.20950714945793153,
      "num_tokens": 27285389.0,
      "step": 14830
    },
    {
      "entropy": 5.195196151733398,
      "epoch": 1.239141329769462,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004850911375507873,
      "loss": 5.0197,
      "mean_token_accuracy": 0.20072472393512725,
      "num_tokens": 27295529.0,
      "step": 14835
    },
    {
      "entropy": 5.250570058822632,
      "epoch": 1.239558970932175,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048508047936467907,
      "loss": 5.0769,
      "mean_token_accuracy": 0.19785389602184295,
      "num_tokens": 27305358.0,
      "step": 14840
    },
    {
      "entropy": 5.188695478439331,
      "epoch": 1.239976612094888,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00048506981750085914,
      "loss": 4.8972,
      "mean_token_accuracy": 0.20335885882377625,
      "num_tokens": 27314765.0,
      "step": 14845
    },
    {
      "entropy": 5.250438499450683,
      "epoch": 1.240394253257601,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00048505915195951423,
      "loss": 4.9746,
      "mean_token_accuracy": 0.20164534002542495,
      "num_tokens": 27324227.0,
      "step": 14850
    },
    {
      "entropy": 5.207560062408447,
      "epoch": 1.2408118944203141,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00048504848274083095,
      "loss": 4.9848,
      "mean_token_accuracy": 0.1973986193537712,
      "num_tokens": 27332890.0,
      "step": 14855
    },
    {
      "entropy": 5.105317163467407,
      "epoch": 1.241229535583027,
      "grad_norm": 1.34375,
      "learning_rate": 0.00048503780984499625,
      "loss": 4.9122,
      "mean_token_accuracy": 0.20852493792772292,
      "num_tokens": 27342411.0,
      "step": 14860
    },
    {
      "entropy": 5.136219453811646,
      "epoch": 1.24164717674574,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004850271332721968,
      "loss": 4.9637,
      "mean_token_accuracy": 0.19931367337703704,
      "num_tokens": 27351649.0,
      "step": 14865
    },
    {
      "entropy": 5.1325160503387455,
      "epoch": 1.242064817908453,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004850164530226196,
      "loss": 4.8793,
      "mean_token_accuracy": 0.20671073496341705,
      "num_tokens": 27360093.0,
      "step": 14870
    },
    {
      "entropy": 5.130922508239746,
      "epoch": 1.242482459071166,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004850057690964515,
      "loss": 4.9241,
      "mean_token_accuracy": 0.2042218342423439,
      "num_tokens": 27368593.0,
      "step": 14875
    },
    {
      "entropy": 5.159080076217651,
      "epoch": 1.242900100233879,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00048499508149387973,
      "loss": 5.02,
      "mean_token_accuracy": 0.18934388607740402,
      "num_tokens": 27378126.0,
      "step": 14880
    },
    {
      "entropy": 5.151203870773315,
      "epoch": 1.2433177413965921,
      "grad_norm": 1.53125,
      "learning_rate": 0.00048498439021509124,
      "loss": 4.8529,
      "mean_token_accuracy": 0.20849937051534653,
      "num_tokens": 27386225.0,
      "step": 14885
    },
    {
      "entropy": 5.214209461212159,
      "epoch": 1.2437353825593052,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004849736952602732,
      "loss": 5.0003,
      "mean_token_accuracy": 0.19747206568717957,
      "num_tokens": 27395629.0,
      "step": 14890
    },
    {
      "entropy": 5.193193149566651,
      "epoch": 1.244153023722018,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000484962996629613,
      "loss": 4.9906,
      "mean_token_accuracy": 0.2096569612622261,
      "num_tokens": 27404431.0,
      "step": 14895
    },
    {
      "entropy": 5.179918670654297,
      "epoch": 1.244570664884731,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048495229432329775,
      "loss": 5.0133,
      "mean_token_accuracy": 0.20366766750812532,
      "num_tokens": 27412944.0,
      "step": 14900
    },
    {
      "entropy": 5.165950584411621,
      "epoch": 1.244988306047444,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00048494158834151485,
      "loss": 4.9207,
      "mean_token_accuracy": 0.19921972006559371,
      "num_tokens": 27422006.0,
      "step": 14905
    },
    {
      "entropy": 5.230476140975952,
      "epoch": 1.245405947210157,
      "grad_norm": 1.328125,
      "learning_rate": 0.00048493087868445176,
      "loss": 4.9939,
      "mean_token_accuracy": 0.20778941214084626,
      "num_tokens": 27431568.0,
      "step": 14910
    },
    {
      "entropy": 5.2268760204315186,
      "epoch": 1.24582358837287,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00048492016535229594,
      "loss": 4.9524,
      "mean_token_accuracy": 0.19952626526355743,
      "num_tokens": 27440471.0,
      "step": 14915
    },
    {
      "entropy": 5.080914783477783,
      "epoch": 1.246241229535583,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004849094483452349,
      "loss": 4.9051,
      "mean_token_accuracy": 0.20287854224443436,
      "num_tokens": 27450209.0,
      "step": 14920
    },
    {
      "entropy": 5.207505941390991,
      "epoch": 1.246658870698296,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004848987276634564,
      "loss": 5.1057,
      "mean_token_accuracy": 0.19572492837905883,
      "num_tokens": 27458876.0,
      "step": 14925
    },
    {
      "entropy": 5.342234802246094,
      "epoch": 1.247076511861009,
      "grad_norm": 1.34375,
      "learning_rate": 0.00048488800330714795,
      "loss": 4.9927,
      "mean_token_accuracy": 0.20349710583686828,
      "num_tokens": 27467526.0,
      "step": 14930
    },
    {
      "entropy": 5.113887214660645,
      "epoch": 1.247494153023722,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00048487727527649733,
      "loss": 4.9177,
      "mean_token_accuracy": 0.20812891870737077,
      "num_tokens": 27477063.0,
      "step": 14935
    },
    {
      "entropy": 5.137446498870849,
      "epoch": 1.247911794186435,
      "grad_norm": 1.359375,
      "learning_rate": 0.00048486654357169246,
      "loss": 5.0313,
      "mean_token_accuracy": 0.19884005934000015,
      "num_tokens": 27486243.0,
      "step": 14940
    },
    {
      "entropy": 5.231842279434204,
      "epoch": 1.248329435349148,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004848558081929211,
      "loss": 5.0072,
      "mean_token_accuracy": 0.19669800996780396,
      "num_tokens": 27495491.0,
      "step": 14945
    },
    {
      "entropy": 5.147360849380493,
      "epoch": 1.2487470765118611,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004848450691403712,
      "loss": 4.9881,
      "mean_token_accuracy": 0.20455342382192612,
      "num_tokens": 27505105.0,
      "step": 14950
    },
    {
      "entropy": 5.154135656356812,
      "epoch": 1.249164717674574,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004848343264142307,
      "loss": 4.9312,
      "mean_token_accuracy": 0.20553508549928665,
      "num_tokens": 27514457.0,
      "step": 14955
    },
    {
      "entropy": 5.281518363952637,
      "epoch": 1.249582358837287,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004848235800146879,
      "loss": 5.0754,
      "mean_token_accuracy": 0.19952864497900008,
      "num_tokens": 27524073.0,
      "step": 14960
    },
    {
      "entropy": 5.1500567436218265,
      "epoch": 1.25,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004848128299419306,
      "loss": 4.9273,
      "mean_token_accuracy": 0.20228756219148636,
      "num_tokens": 27532840.0,
      "step": 14965
    },
    {
      "entropy": 5.232335758209229,
      "epoch": 1.250417641162713,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00048480207619614726,
      "loss": 5.0769,
      "mean_token_accuracy": 0.19289935231208802,
      "num_tokens": 27541775.0,
      "step": 14970
    },
    {
      "entropy": 5.2373046875,
      "epoch": 1.250835282325426,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000484791318777526,
      "loss": 4.976,
      "mean_token_accuracy": 0.20685369968414308,
      "num_tokens": 27550668.0,
      "step": 14975
    },
    {
      "entropy": 5.165564250946045,
      "epoch": 1.2512529234881389,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00048478055768625517,
      "loss": 5.0098,
      "mean_token_accuracy": 0.20409852117300034,
      "num_tokens": 27560255.0,
      "step": 14980
    },
    {
      "entropy": 5.112481498718262,
      "epoch": 1.251670564650852,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048476979292252323,
      "loss": 4.9404,
      "mean_token_accuracy": 0.19943113923072814,
      "num_tokens": 27569028.0,
      "step": 14985
    },
    {
      "entropy": 5.175985717773438,
      "epoch": 1.252088205813565,
      "grad_norm": 1.25,
      "learning_rate": 0.0004847590244865185,
      "loss": 4.8654,
      "mean_token_accuracy": 0.20914251804351808,
      "num_tokens": 27577396.0,
      "step": 14990
    },
    {
      "entropy": 5.177746677398682,
      "epoch": 1.252505846976278,
      "grad_norm": 1.46875,
      "learning_rate": 0.00048474825237842964,
      "loss": 5.0254,
      "mean_token_accuracy": 0.1929772228002548,
      "num_tokens": 27586710.0,
      "step": 14995
    },
    {
      "entropy": 5.10919041633606,
      "epoch": 1.252923488138991,
      "grad_norm": 1.515625,
      "learning_rate": 0.000484737476598445,
      "loss": 4.8778,
      "mean_token_accuracy": 0.21211249083280564,
      "num_tokens": 27594842.0,
      "step": 15000
    },
    {
      "epoch": 1.252923488138991,
      "eval_entropy": 4.920714459967699,
      "eval_loss": 5.014506816864014,
      "eval_mean_token_accuracy": 0.21129573983576458,
      "eval_num_tokens": 27594842.0,
      "eval_runtime": 27.6198,
      "eval_samples_per_second": 1367.245,
      "eval_steps_per_second": 170.928,
      "step": 15000
    },
    {
      "entropy": 5.0056804656982425,
      "epoch": 1.253341129301704,
      "grad_norm": 1.296875,
      "learning_rate": 0.00048472669714675357,
      "loss": 4.8729,
      "mean_token_accuracy": 0.20810110867023468,
      "num_tokens": 27604661.0,
      "step": 15005
    },
    {
      "entropy": 5.086064052581787,
      "epoch": 1.253758770464417,
      "grad_norm": 1.359375,
      "learning_rate": 0.00048471591402354374,
      "loss": 4.815,
      "mean_token_accuracy": 0.21455551385879518,
      "num_tokens": 27614015.0,
      "step": 15010
    },
    {
      "entropy": 5.2216033935546875,
      "epoch": 1.25417641162713,
      "grad_norm": 1.25,
      "learning_rate": 0.00048470512722900434,
      "loss": 5.0741,
      "mean_token_accuracy": 0.19608521610498428,
      "num_tokens": 27624164.0,
      "step": 15015
    },
    {
      "entropy": 5.191699647903443,
      "epoch": 1.254594052789843,
      "grad_norm": 1.375,
      "learning_rate": 0.00048469433676332434,
      "loss": 4.9424,
      "mean_token_accuracy": 0.20993922501802445,
      "num_tokens": 27633993.0,
      "step": 15020
    },
    {
      "entropy": 5.189644622802734,
      "epoch": 1.255011693952556,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004846835426266926,
      "loss": 4.983,
      "mean_token_accuracy": 0.20420642495155333,
      "num_tokens": 27643772.0,
      "step": 15025
    },
    {
      "entropy": 5.178664112091065,
      "epoch": 1.255429335115269,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004846727448192981,
      "loss": 4.8916,
      "mean_token_accuracy": 0.20651845633983612,
      "num_tokens": 27652511.0,
      "step": 15030
    },
    {
      "entropy": 5.116759872436523,
      "epoch": 1.255846976277982,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00048466194334132964,
      "loss": 4.8846,
      "mean_token_accuracy": 0.20806932747364043,
      "num_tokens": 27661589.0,
      "step": 15035
    },
    {
      "entropy": 5.081556367874145,
      "epoch": 1.2562646174406948,
      "grad_norm": 1.328125,
      "learning_rate": 0.00048465113819297654,
      "loss": 4.9618,
      "mean_token_accuracy": 0.20809750407934188,
      "num_tokens": 27670299.0,
      "step": 15040
    },
    {
      "entropy": 5.269770669937134,
      "epoch": 1.2566822586034079,
      "grad_norm": 2.25,
      "learning_rate": 0.000484640329374428,
      "loss": 5.088,
      "mean_token_accuracy": 0.19591086208820344,
      "num_tokens": 27679375.0,
      "step": 15045
    },
    {
      "entropy": 5.1649175643920895,
      "epoch": 1.257099899766121,
      "grad_norm": 1.296875,
      "learning_rate": 0.00048462951688587317,
      "loss": 4.939,
      "mean_token_accuracy": 0.2154373750090599,
      "num_tokens": 27688754.0,
      "step": 15050
    },
    {
      "entropy": 5.105102300643921,
      "epoch": 1.257517540928834,
      "grad_norm": 1.296875,
      "learning_rate": 0.00048461870072750117,
      "loss": 4.9709,
      "mean_token_accuracy": 0.1949514552950859,
      "num_tokens": 27697799.0,
      "step": 15055
    },
    {
      "entropy": 5.126546382904053,
      "epoch": 1.257935182091547,
      "grad_norm": 1.453125,
      "learning_rate": 0.00048460788089950164,
      "loss": 4.8966,
      "mean_token_accuracy": 0.21081169098615646,
      "num_tokens": 27707239.0,
      "step": 15060
    },
    {
      "entropy": 5.201789855957031,
      "epoch": 1.25835282325426,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004845970574020637,
      "loss": 4.9503,
      "mean_token_accuracy": 0.2030317395925522,
      "num_tokens": 27716027.0,
      "step": 15065
    },
    {
      "entropy": 5.1384024143219,
      "epoch": 1.258770464416973,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004845862302353771,
      "loss": 4.8422,
      "mean_token_accuracy": 0.21493653655052186,
      "num_tokens": 27724849.0,
      "step": 15070
    },
    {
      "entropy": 5.095423173904419,
      "epoch": 1.2591881055796859,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00048457539939963126,
      "loss": 4.8706,
      "mean_token_accuracy": 0.20886123180389404,
      "num_tokens": 27733899.0,
      "step": 15075
    },
    {
      "entropy": 5.133898067474365,
      "epoch": 1.259605746742399,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00048456456489501574,
      "loss": 5.0101,
      "mean_token_accuracy": 0.20155577957630158,
      "num_tokens": 27743967.0,
      "step": 15080
    },
    {
      "entropy": 5.173690509796143,
      "epoch": 1.260023387905112,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004845537267217202,
      "loss": 4.9393,
      "mean_token_accuracy": 0.20087562054395675,
      "num_tokens": 27753105.0,
      "step": 15085
    },
    {
      "entropy": 5.23564453125,
      "epoch": 1.260441029067825,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004845428848799346,
      "loss": 5.0333,
      "mean_token_accuracy": 0.20349034518003464,
      "num_tokens": 27762771.0,
      "step": 15090
    },
    {
      "entropy": 5.1439300060272215,
      "epoch": 1.260858670230538,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004845320393698484,
      "loss": 4.9196,
      "mean_token_accuracy": 0.2094682902097702,
      "num_tokens": 27771653.0,
      "step": 15095
    },
    {
      "entropy": 5.109306764602661,
      "epoch": 1.2612763113932508,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004845211901916517,
      "loss": 4.9827,
      "mean_token_accuracy": 0.2039654940366745,
      "num_tokens": 27780585.0,
      "step": 15100
    },
    {
      "entropy": 5.125766706466675,
      "epoch": 1.2616939525559638,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00048451033734553444,
      "loss": 4.9256,
      "mean_token_accuracy": 0.20397261381149293,
      "num_tokens": 27790625.0,
      "step": 15105
    },
    {
      "entropy": 5.375638151168824,
      "epoch": 1.2621115937186769,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004844994808316864,
      "loss": 5.1987,
      "mean_token_accuracy": 0.19191684424877167,
      "num_tokens": 27800948.0,
      "step": 15110
    },
    {
      "entropy": 5.265510129928589,
      "epoch": 1.26252923488139,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004844886206502979,
      "loss": 5.057,
      "mean_token_accuracy": 0.1997806191444397,
      "num_tokens": 27809973.0,
      "step": 15115
    },
    {
      "entropy": 5.192895174026489,
      "epoch": 1.262946876044103,
      "grad_norm": 1.375,
      "learning_rate": 0.00048447775680155884,
      "loss": 4.9595,
      "mean_token_accuracy": 0.20437342077493667,
      "num_tokens": 27818738.0,
      "step": 15120
    },
    {
      "entropy": 5.118112373352051,
      "epoch": 1.263364517206816,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00048446688928565956,
      "loss": 4.9308,
      "mean_token_accuracy": 0.21279027462005615,
      "num_tokens": 27827537.0,
      "step": 15125
    },
    {
      "entropy": 5.3163567066192625,
      "epoch": 1.263782158369529,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004844560181027903,
      "loss": 5.1059,
      "mean_token_accuracy": 0.19123855084180832,
      "num_tokens": 27837498.0,
      "step": 15130
    },
    {
      "entropy": 5.231694507598877,
      "epoch": 1.2641997995322418,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004844451432531412,
      "loss": 4.988,
      "mean_token_accuracy": 0.19799877256155013,
      "num_tokens": 27847547.0,
      "step": 15135
    },
    {
      "entropy": 5.163742780685425,
      "epoch": 1.2646174406949549,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004844342647369028,
      "loss": 4.9308,
      "mean_token_accuracy": 0.2078368231654167,
      "num_tokens": 27856468.0,
      "step": 15140
    },
    {
      "entropy": 5.231487560272217,
      "epoch": 1.2650350818576679,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00048442338255426555,
      "loss": 5.0999,
      "mean_token_accuracy": 0.19490432739257812,
      "num_tokens": 27865324.0,
      "step": 15145
    },
    {
      "entropy": 5.14596791267395,
      "epoch": 1.265452723020381,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004844124967054199,
      "loss": 4.9692,
      "mean_token_accuracy": 0.20157322734594346,
      "num_tokens": 27875364.0,
      "step": 15150
    },
    {
      "entropy": 5.215729570388794,
      "epoch": 1.265870364183094,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004844016071905564,
      "loss": 4.9388,
      "mean_token_accuracy": 0.20703227669000626,
      "num_tokens": 27884739.0,
      "step": 15155
    },
    {
      "entropy": 5.067172384262085,
      "epoch": 1.2662880053458068,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004843907140098656,
      "loss": 4.8367,
      "mean_token_accuracy": 0.2161872446537018,
      "num_tokens": 27893235.0,
      "step": 15160
    },
    {
      "entropy": 5.091639518737793,
      "epoch": 1.2667056465085198,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004843798171635384,
      "loss": 4.9904,
      "mean_token_accuracy": 0.20268763452768326,
      "num_tokens": 27902090.0,
      "step": 15165
    },
    {
      "entropy": 5.201542568206787,
      "epoch": 1.2671232876712328,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004843689166517655,
      "loss": 4.9391,
      "mean_token_accuracy": 0.20061081349849702,
      "num_tokens": 27910702.0,
      "step": 15170
    },
    {
      "entropy": 5.19600920677185,
      "epoch": 1.2675409288339459,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004843580124747377,
      "loss": 4.9744,
      "mean_token_accuracy": 0.20476946979761124,
      "num_tokens": 27919296.0,
      "step": 15175
    },
    {
      "entropy": 5.158300256729126,
      "epoch": 1.267958569996659,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00048434710463264585,
      "loss": 4.9685,
      "mean_token_accuracy": 0.20941117107868196,
      "num_tokens": 27929074.0,
      "step": 15180
    },
    {
      "entropy": 5.223547172546387,
      "epoch": 1.268376211159372,
      "grad_norm": 1.25,
      "learning_rate": 0.00048433619312568095,
      "loss": 5.0341,
      "mean_token_accuracy": 0.19935351610183716,
      "num_tokens": 27938227.0,
      "step": 15185
    },
    {
      "entropy": 5.264383268356323,
      "epoch": 1.268793852322085,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00048432527795403405,
      "loss": 5.0171,
      "mean_token_accuracy": 0.19871484637260436,
      "num_tokens": 27948050.0,
      "step": 15190
    },
    {
      "entropy": 5.186510133743286,
      "epoch": 1.2692114934847978,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004843143591178962,
      "loss": 4.8929,
      "mean_token_accuracy": 0.20017318725585936,
      "num_tokens": 27957702.0,
      "step": 15195
    },
    {
      "entropy": 5.211152982711792,
      "epoch": 1.2696291346475108,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00048430343661745844,
      "loss": 5.0575,
      "mean_token_accuracy": 0.1936388745903969,
      "num_tokens": 27966085.0,
      "step": 15200
    },
    {
      "entropy": 5.210051822662353,
      "epoch": 1.2700467758102238,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004842925104529122,
      "loss": 5.0509,
      "mean_token_accuracy": 0.19898343235254287,
      "num_tokens": 27976228.0,
      "step": 15205
    },
    {
      "entropy": 5.227244758605957,
      "epoch": 1.2704644169729369,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00048428158062444857,
      "loss": 4.9669,
      "mean_token_accuracy": 0.20769187957048416,
      "num_tokens": 27984910.0,
      "step": 15210
    },
    {
      "entropy": 5.209272050857544,
      "epoch": 1.27088205813565,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048427064713225894,
      "loss": 4.9539,
      "mean_token_accuracy": 0.20423158556222915,
      "num_tokens": 27993767.0,
      "step": 15215
    },
    {
      "entropy": 5.194864749908447,
      "epoch": 1.2712996992983627,
      "grad_norm": 1.796875,
      "learning_rate": 0.00048425970997653477,
      "loss": 4.9752,
      "mean_token_accuracy": 0.2035406932234764,
      "num_tokens": 28002587.0,
      "step": 15220
    },
    {
      "entropy": 5.162222814559937,
      "epoch": 1.2717173404610758,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004842487691574675,
      "loss": 5.019,
      "mean_token_accuracy": 0.19605301320552826,
      "num_tokens": 28011598.0,
      "step": 15225
    },
    {
      "entropy": 5.240627813339233,
      "epoch": 1.2721349816237888,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00048423782467524865,
      "loss": 5.0945,
      "mean_token_accuracy": 0.2005903825163841,
      "num_tokens": 28021321.0,
      "step": 15230
    },
    {
      "entropy": 5.237555313110351,
      "epoch": 1.2725526227865018,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004842268765300698,
      "loss": 5.0044,
      "mean_token_accuracy": 0.2054242491722107,
      "num_tokens": 28031009.0,
      "step": 15235
    },
    {
      "entropy": 5.227293348312378,
      "epoch": 1.2729702639492149,
      "grad_norm": 1.375,
      "learning_rate": 0.00048421592472212263,
      "loss": 5.0401,
      "mean_token_accuracy": 0.19711171686649323,
      "num_tokens": 28040330.0,
      "step": 15240
    },
    {
      "entropy": 5.141539669036865,
      "epoch": 1.273387905111928,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00048420496925159886,
      "loss": 5.0021,
      "mean_token_accuracy": 0.19651335030794143,
      "num_tokens": 28048952.0,
      "step": 15245
    },
    {
      "entropy": 5.19468731880188,
      "epoch": 1.273805546274641,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004841940101186903,
      "loss": 4.9705,
      "mean_token_accuracy": 0.19303844571113588,
      "num_tokens": 28058803.0,
      "step": 15250
    },
    {
      "entropy": 5.222444772720337,
      "epoch": 1.2742231874373537,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004841830473235888,
      "loss": 4.9637,
      "mean_token_accuracy": 0.20900567024946212,
      "num_tokens": 28067978.0,
      "step": 15255
    },
    {
      "entropy": 5.081825113296508,
      "epoch": 1.2746408286000668,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00048417208086648625,
      "loss": 4.9297,
      "mean_token_accuracy": 0.20809237509965897,
      "num_tokens": 28077229.0,
      "step": 15260
    },
    {
      "entropy": 5.179949426651001,
      "epoch": 1.2750584697627798,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004841611107475746,
      "loss": 4.9637,
      "mean_token_accuracy": 0.2089214339852333,
      "num_tokens": 28086316.0,
      "step": 15265
    },
    {
      "entropy": 5.334323406219482,
      "epoch": 1.2754761109254928,
      "grad_norm": 1.375,
      "learning_rate": 0.000484150136967046,
      "loss": 5.1488,
      "mean_token_accuracy": 0.19488464295864105,
      "num_tokens": 28096914.0,
      "step": 15270
    },
    {
      "entropy": 5.222511100769043,
      "epoch": 1.2758937520882059,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004841391595250925,
      "loss": 5.0249,
      "mean_token_accuracy": 0.1960061401128769,
      "num_tokens": 28106532.0,
      "step": 15275
    },
    {
      "entropy": 5.191496896743774,
      "epoch": 1.2763113932509187,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004841281784219063,
      "loss": 4.9855,
      "mean_token_accuracy": 0.20173507928848267,
      "num_tokens": 28115052.0,
      "step": 15280
    },
    {
      "entropy": 5.175038146972656,
      "epoch": 1.2767290344136317,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00048411719365767947,
      "loss": 4.9016,
      "mean_token_accuracy": 0.21106401681900025,
      "num_tokens": 28123571.0,
      "step": 15285
    },
    {
      "entropy": 5.075457954406739,
      "epoch": 1.2771466755763448,
      "grad_norm": 1.21875,
      "learning_rate": 0.00048410620523260455,
      "loss": 4.8471,
      "mean_token_accuracy": 0.21511422842741013,
      "num_tokens": 28131973.0,
      "step": 15290
    },
    {
      "entropy": 5.17520923614502,
      "epoch": 1.2775643167390578,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048409521314687383,
      "loss": 5.0744,
      "mean_token_accuracy": 0.1930087000131607,
      "num_tokens": 28140965.0,
      "step": 15295
    },
    {
      "entropy": 5.202703046798706,
      "epoch": 1.2779819579017708,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004840842174006797,
      "loss": 5.0329,
      "mean_token_accuracy": 0.2008312985301018,
      "num_tokens": 28150652.0,
      "step": 15300
    },
    {
      "entropy": 5.364263010025025,
      "epoch": 1.2783995990644839,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004840732179942146,
      "loss": 5.2189,
      "mean_token_accuracy": 0.18567581921815873,
      "num_tokens": 28160830.0,
      "step": 15305
    },
    {
      "entropy": 5.277627897262573,
      "epoch": 1.278817240227197,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004840622149276712,
      "loss": 4.8535,
      "mean_token_accuracy": 0.20971939861774444,
      "num_tokens": 28169423.0,
      "step": 15310
    },
    {
      "entropy": 5.156337833404541,
      "epoch": 1.2792348813899097,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004840512082012421,
      "loss": 4.9382,
      "mean_token_accuracy": 0.2027086317539215,
      "num_tokens": 28178997.0,
      "step": 15315
    },
    {
      "entropy": 5.134846353530884,
      "epoch": 1.2796525225526227,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00048404019781511995,
      "loss": 4.9499,
      "mean_token_accuracy": 0.2064346581697464,
      "num_tokens": 28188014.0,
      "step": 15320
    },
    {
      "entropy": 5.206334638595581,
      "epoch": 1.2800701637153358,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048402918376949753,
      "loss": 5.0186,
      "mean_token_accuracy": 0.19418144822120667,
      "num_tokens": 28197800.0,
      "step": 15325
    },
    {
      "entropy": 5.197379684448242,
      "epoch": 1.2804878048780488,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004840181660645676,
      "loss": 4.9126,
      "mean_token_accuracy": 0.20897612124681472,
      "num_tokens": 28207388.0,
      "step": 15330
    },
    {
      "entropy": 5.178161525726319,
      "epoch": 1.2809054460407618,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004840071447005231,
      "loss": 4.9766,
      "mean_token_accuracy": 0.2100621849298477,
      "num_tokens": 28216774.0,
      "step": 15335
    },
    {
      "entropy": 5.187810039520263,
      "epoch": 1.2813230872034747,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000483996119677557,
      "loss": 4.993,
      "mean_token_accuracy": 0.2028380438685417,
      "num_tokens": 28226402.0,
      "step": 15340
    },
    {
      "entropy": 5.171538925170898,
      "epoch": 1.281740728366188,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00048398509099586225,
      "loss": 4.9939,
      "mean_token_accuracy": 0.19431959390640258,
      "num_tokens": 28235890.0,
      "step": 15345
    },
    {
      "entropy": 5.166479444503784,
      "epoch": 1.2821583695289007,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00048397405865563186,
      "loss": 4.9406,
      "mean_token_accuracy": 0.2082495555281639,
      "num_tokens": 28244121.0,
      "step": 15350
    },
    {
      "entropy": 5.261213397979736,
      "epoch": 1.2825760106916138,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004839630226570591,
      "loss": 5.0292,
      "mean_token_accuracy": 0.19818469136953354,
      "num_tokens": 28252259.0,
      "step": 15355
    },
    {
      "entropy": 5.130181360244751,
      "epoch": 1.2829936518543268,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000483951983000337,
      "loss": 4.8861,
      "mean_token_accuracy": 0.2094778060913086,
      "num_tokens": 28262199.0,
      "step": 15360
    },
    {
      "entropy": 5.17686276435852,
      "epoch": 1.2834112930170398,
      "grad_norm": 1.296875,
      "learning_rate": 0.000483940939685659,
      "loss": 4.9796,
      "mean_token_accuracy": 0.20316151827573775,
      "num_tokens": 28271120.0,
      "step": 15365
    },
    {
      "entropy": 5.218368291854858,
      "epoch": 1.2838289341797529,
      "grad_norm": 1.40625,
      "learning_rate": 0.00048392989271321836,
      "loss": 4.961,
      "mean_token_accuracy": 0.20530664622783662,
      "num_tokens": 28279766.0,
      "step": 15370
    },
    {
      "entropy": 5.189590215682983,
      "epoch": 1.2842465753424657,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004839188420832084,
      "loss": 4.9681,
      "mean_token_accuracy": 0.20657599121332168,
      "num_tokens": 28288872.0,
      "step": 15375
    },
    {
      "entropy": 5.091854476928711,
      "epoch": 1.2846642165051787,
      "grad_norm": 1.3125,
      "learning_rate": 0.00048390778779582266,
      "loss": 4.9169,
      "mean_token_accuracy": 0.20289535969495773,
      "num_tokens": 28297719.0,
      "step": 15380
    },
    {
      "entropy": 5.19854588508606,
      "epoch": 1.2850818576678917,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00048389672985125467,
      "loss": 5.0404,
      "mean_token_accuracy": 0.1983731895685196,
      "num_tokens": 28306547.0,
      "step": 15385
    },
    {
      "entropy": 5.158060884475708,
      "epoch": 1.2854994988306048,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004838856682496979,
      "loss": 4.9675,
      "mean_token_accuracy": 0.1996526151895523,
      "num_tokens": 28316387.0,
      "step": 15390
    },
    {
      "entropy": 5.2526342391967775,
      "epoch": 1.2859171399933178,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004838746029913461,
      "loss": 5.0433,
      "mean_token_accuracy": 0.19401377737522124,
      "num_tokens": 28326642.0,
      "step": 15395
    },
    {
      "entropy": 5.177738046646118,
      "epoch": 1.2863347811560306,
      "grad_norm": 1.21875,
      "learning_rate": 0.000483863534076393,
      "loss": 4.9368,
      "mean_token_accuracy": 0.20567625463008882,
      "num_tokens": 28335647.0,
      "step": 15400
    },
    {
      "entropy": 5.109444856643677,
      "epoch": 1.2867524223187439,
      "grad_norm": 1.25,
      "learning_rate": 0.0004838524615050323,
      "loss": 4.973,
      "mean_token_accuracy": 0.20189628452062608,
      "num_tokens": 28344536.0,
      "step": 15405
    },
    {
      "entropy": 5.185155057907105,
      "epoch": 1.2871700634814567,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004838413852774579,
      "loss": 4.982,
      "mean_token_accuracy": 0.20259611457586288,
      "num_tokens": 28354121.0,
      "step": 15410
    },
    {
      "entropy": 5.117042446136475,
      "epoch": 1.2875877046441697,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00048383030539386355,
      "loss": 4.9394,
      "mean_token_accuracy": 0.2071150556206703,
      "num_tokens": 28362215.0,
      "step": 15415
    },
    {
      "entropy": 5.223753261566162,
      "epoch": 1.2880053458068828,
      "grad_norm": 1.5,
      "learning_rate": 0.0004838192218544434,
      "loss": 5.0012,
      "mean_token_accuracy": 0.1995368257164955,
      "num_tokens": 28371323.0,
      "step": 15420
    },
    {
      "entropy": 5.293983602523804,
      "epoch": 1.2884229869695958,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004838081346593915,
      "loss": 5.0571,
      "mean_token_accuracy": 0.19423929005861282,
      "num_tokens": 28380333.0,
      "step": 15425
    },
    {
      "entropy": 5.19257493019104,
      "epoch": 1.2888406281323088,
      "grad_norm": 1.296875,
      "learning_rate": 0.00048379704380890177,
      "loss": 4.9697,
      "mean_token_accuracy": 0.19789639264345169,
      "num_tokens": 28389021.0,
      "step": 15430
    },
    {
      "entropy": 5.130419445037842,
      "epoch": 1.2892582692950216,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004837859493031685,
      "loss": 4.8885,
      "mean_token_accuracy": 0.1992548882961273,
      "num_tokens": 28398309.0,
      "step": 15435
    },
    {
      "entropy": 5.079394435882568,
      "epoch": 1.2896759104577347,
      "grad_norm": 1.421875,
      "learning_rate": 0.00048377485114238576,
      "loss": 4.8079,
      "mean_token_accuracy": 0.21296215057373047,
      "num_tokens": 28406946.0,
      "step": 15440
    },
    {
      "entropy": 5.154804849624634,
      "epoch": 1.2900935516204477,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004837637493267481,
      "loss": 4.9671,
      "mean_token_accuracy": 0.20934919267892838,
      "num_tokens": 28416561.0,
      "step": 15445
    },
    {
      "entropy": 5.116442251205444,
      "epoch": 1.2905111927831607,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004837526438564496,
      "loss": 4.8097,
      "mean_token_accuracy": 0.22329769134521485,
      "num_tokens": 28425133.0,
      "step": 15450
    },
    {
      "entropy": 5.139532995223999,
      "epoch": 1.2909288339458738,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004837415347316849,
      "loss": 5.0169,
      "mean_token_accuracy": 0.19733298271894456,
      "num_tokens": 28434282.0,
      "step": 15455
    },
    {
      "entropy": 5.2291747570037845,
      "epoch": 1.2913464751085866,
      "grad_norm": 1.4375,
      "learning_rate": 0.00048373042195264836,
      "loss": 5.0061,
      "mean_token_accuracy": 0.19552657306194304,
      "num_tokens": 28443149.0,
      "step": 15460
    },
    {
      "entropy": 5.262494945526123,
      "epoch": 1.2917641162712998,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004837193055195344,
      "loss": 5.0625,
      "mean_token_accuracy": 0.1947874903678894,
      "num_tokens": 28453294.0,
      "step": 15465
    },
    {
      "entropy": 5.160983371734619,
      "epoch": 1.2921817574340126,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00048370818543253793,
      "loss": 4.9024,
      "mean_token_accuracy": 0.21207103729248047,
      "num_tokens": 28462518.0,
      "step": 15470
    },
    {
      "entropy": 5.172740697860718,
      "epoch": 1.2925993985967257,
      "grad_norm": 1.4375,
      "learning_rate": 0.00048369706169185334,
      "loss": 4.9826,
      "mean_token_accuracy": 0.20513939261436462,
      "num_tokens": 28470674.0,
      "step": 15475
    },
    {
      "entropy": 5.1887115955352785,
      "epoch": 1.2930170397594387,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00048368593429767555,
      "loss": 4.9962,
      "mean_token_accuracy": 0.20158345848321915,
      "num_tokens": 28480088.0,
      "step": 15480
    },
    {
      "entropy": 5.236731052398682,
      "epoch": 1.2934346809221517,
      "grad_norm": 1.28125,
      "learning_rate": 0.00048367480325019923,
      "loss": 5.0361,
      "mean_token_accuracy": 0.19880567491054535,
      "num_tokens": 28489533.0,
      "step": 15485
    },
    {
      "entropy": 5.239792346954346,
      "epoch": 1.2938523220848648,
      "grad_norm": 1.171875,
      "learning_rate": 0.00048366366854961937,
      "loss": 4.9866,
      "mean_token_accuracy": 0.20567111372947694,
      "num_tokens": 28498758.0,
      "step": 15490
    },
    {
      "entropy": 5.135676717758178,
      "epoch": 1.2942699632475776,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004836525301961307,
      "loss": 4.9311,
      "mean_token_accuracy": 0.20519788563251495,
      "num_tokens": 28507633.0,
      "step": 15495
    },
    {
      "entropy": 5.111695575714111,
      "epoch": 1.2946876044102906,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004836413881899285,
      "loss": 4.9442,
      "mean_token_accuracy": 0.20441281348466872,
      "num_tokens": 28516905.0,
      "step": 15500
    },
    {
      "entropy": 5.189840698242188,
      "epoch": 1.2951052455730037,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004836302425312075,
      "loss": 5.0623,
      "mean_token_accuracy": 0.19709089994430543,
      "num_tokens": 28527419.0,
      "step": 15505
    },
    {
      "entropy": 5.097623300552368,
      "epoch": 1.2955228867357167,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004836190932201631,
      "loss": 4.7981,
      "mean_token_accuracy": 0.21490249633789063,
      "num_tokens": 28535950.0,
      "step": 15510
    },
    {
      "entropy": 5.160767555236816,
      "epoch": 1.2959405278984297,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00048360794025699026,
      "loss": 4.8921,
      "mean_token_accuracy": 0.2086631804704666,
      "num_tokens": 28544341.0,
      "step": 15515
    },
    {
      "entropy": 5.23278112411499,
      "epoch": 1.2963581690611425,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004835967836418844,
      "loss": 5.0185,
      "mean_token_accuracy": 0.19932279586791993,
      "num_tokens": 28553805.0,
      "step": 15520
    },
    {
      "entropy": 5.201968097686768,
      "epoch": 1.2967758102238558,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004835856233750407,
      "loss": 4.9185,
      "mean_token_accuracy": 0.20457566529512405,
      "num_tokens": 28562224.0,
      "step": 15525
    },
    {
      "entropy": 5.096701622009277,
      "epoch": 1.2971934513865686,
      "grad_norm": 1.203125,
      "learning_rate": 0.00048357445945665465,
      "loss": 4.8443,
      "mean_token_accuracy": 0.21654272377490996,
      "num_tokens": 28570771.0,
      "step": 15530
    },
    {
      "entropy": 5.215668106079102,
      "epoch": 1.2976110925492816,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004835632918869215,
      "loss": 5.063,
      "mean_token_accuracy": 0.19906616657972337,
      "num_tokens": 28580447.0,
      "step": 15535
    },
    {
      "entropy": 5.322928428649902,
      "epoch": 1.2980287337119947,
      "grad_norm": 1.1875,
      "learning_rate": 0.000483552120666037,
      "loss": 5.0985,
      "mean_token_accuracy": 0.19962597340345384,
      "num_tokens": 28590472.0,
      "step": 15540
    },
    {
      "entropy": 5.186050271987915,
      "epoch": 1.2984463748747077,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048354094579419656,
      "loss": 4.9578,
      "mean_token_accuracy": 0.20493223518133163,
      "num_tokens": 28599201.0,
      "step": 15545
    },
    {
      "entropy": 5.138039922714233,
      "epoch": 1.2988640160374207,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004835297672715958,
      "loss": 4.9825,
      "mean_token_accuracy": 0.19557948857545854,
      "num_tokens": 28608499.0,
      "step": 15550
    },
    {
      "entropy": 5.208690977096557,
      "epoch": 1.2992816572001336,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00048351858509843054,
      "loss": 4.9433,
      "mean_token_accuracy": 0.2075222462415695,
      "num_tokens": 28618310.0,
      "step": 15555
    },
    {
      "entropy": 5.238944864273071,
      "epoch": 1.2996992983628466,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004835073992748964,
      "loss": 4.9791,
      "mean_token_accuracy": 0.19718073308467865,
      "num_tokens": 28627374.0,
      "step": 15560
    },
    {
      "entropy": 5.156641817092895,
      "epoch": 1.3001169395255596,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00048349620980118926,
      "loss": 4.9404,
      "mean_token_accuracy": 0.19804110676050185,
      "num_tokens": 28636331.0,
      "step": 15565
    },
    {
      "entropy": 5.15984206199646,
      "epoch": 1.3005345806882727,
      "grad_norm": 1.28125,
      "learning_rate": 0.000483485016677505,
      "loss": 4.9183,
      "mean_token_accuracy": 0.20359799712896348,
      "num_tokens": 28645006.0,
      "step": 15570
    },
    {
      "entropy": 5.139520978927612,
      "epoch": 1.3009522218509857,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00048347381990403957,
      "loss": 4.963,
      "mean_token_accuracy": 0.2064681813120842,
      "num_tokens": 28653202.0,
      "step": 15575
    },
    {
      "entropy": 5.168787097930908,
      "epoch": 1.3013698630136985,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00048346261948098895,
      "loss": 4.9612,
      "mean_token_accuracy": 0.20327424108982087,
      "num_tokens": 28662701.0,
      "step": 15580
    },
    {
      "entropy": 5.20242748260498,
      "epoch": 1.3017875041764118,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00048345141540854933,
      "loss": 4.959,
      "mean_token_accuracy": 0.20309157967567443,
      "num_tokens": 28670967.0,
      "step": 15585
    },
    {
      "entropy": 5.227055454254151,
      "epoch": 1.3022051453391246,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00048344020768691673,
      "loss": 5.0253,
      "mean_token_accuracy": 0.20532655119895935,
      "num_tokens": 28679458.0,
      "step": 15590
    },
    {
      "entropy": 5.156626033782959,
      "epoch": 1.3026227865018376,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004834289963162874,
      "loss": 4.902,
      "mean_token_accuracy": 0.21451759934425355,
      "num_tokens": 28688497.0,
      "step": 15595
    },
    {
      "entropy": 5.154309415817261,
      "epoch": 1.3030404276645506,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004834177812968576,
      "loss": 4.9588,
      "mean_token_accuracy": 0.19972556382417678,
      "num_tokens": 28697729.0,
      "step": 15600
    },
    {
      "entropy": 5.267010688781738,
      "epoch": 1.3034580688272637,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048340656262882366,
      "loss": 5.0626,
      "mean_token_accuracy": 0.19472329914569855,
      "num_tokens": 28707778.0,
      "step": 15605
    },
    {
      "entropy": 5.215834379196167,
      "epoch": 1.3038757099899767,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00048339534031238197,
      "loss": 5.0207,
      "mean_token_accuracy": 0.1989232987165451,
      "num_tokens": 28718010.0,
      "step": 15610
    },
    {
      "entropy": 5.258842849731446,
      "epoch": 1.3042933511526895,
      "grad_norm": 1.234375,
      "learning_rate": 0.000483384114347729,
      "loss": 4.9471,
      "mean_token_accuracy": 0.2049952119588852,
      "num_tokens": 28727279.0,
      "step": 15615
    },
    {
      "entropy": 5.170647048950196,
      "epoch": 1.3047109923154026,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004833728847350613,
      "loss": 5.0177,
      "mean_token_accuracy": 0.20353060960769653,
      "num_tokens": 28736510.0,
      "step": 15620
    },
    {
      "entropy": 5.186431884765625,
      "epoch": 1.3051286334781156,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00048336165147457537,
      "loss": 4.9851,
      "mean_token_accuracy": 0.20114415436983107,
      "num_tokens": 28745226.0,
      "step": 15625
    },
    {
      "entropy": 5.197024869918823,
      "epoch": 1.3055462746408286,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004833504145664681,
      "loss": 4.9509,
      "mean_token_accuracy": 0.20538980960845948,
      "num_tokens": 28754052.0,
      "step": 15630
    },
    {
      "entropy": 5.171851539611817,
      "epoch": 1.3059639158035417,
      "grad_norm": 1.40625,
      "learning_rate": 0.00048333917401093587,
      "loss": 4.9152,
      "mean_token_accuracy": 0.20969721674919128,
      "num_tokens": 28763580.0,
      "step": 15635
    },
    {
      "entropy": 5.193964290618896,
      "epoch": 1.3063815569662545,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00048332792980817563,
      "loss": 4.9256,
      "mean_token_accuracy": 0.2025534391403198,
      "num_tokens": 28772540.0,
      "step": 15640
    },
    {
      "entropy": 5.170376586914062,
      "epoch": 1.3067991981289677,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004833166819583843,
      "loss": 4.9064,
      "mean_token_accuracy": 0.20521347522735595,
      "num_tokens": 28782027.0,
      "step": 15645
    },
    {
      "entropy": 5.188032341003418,
      "epoch": 1.3072168392916805,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004833054304617586,
      "loss": 4.9487,
      "mean_token_accuracy": 0.20348575860261917,
      "num_tokens": 28791724.0,
      "step": 15650
    },
    {
      "entropy": 5.175020599365235,
      "epoch": 1.3076344804543936,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00048329417531849567,
      "loss": 4.9952,
      "mean_token_accuracy": 0.20600388050079346,
      "num_tokens": 28800681.0,
      "step": 15655
    },
    {
      "entropy": 5.163554906845093,
      "epoch": 1.3080521216171066,
      "grad_norm": 1.421875,
      "learning_rate": 0.00048328291652879245,
      "loss": 5.0267,
      "mean_token_accuracy": 0.20625173300504684,
      "num_tokens": 28810121.0,
      "step": 15660
    },
    {
      "entropy": 5.160272645950317,
      "epoch": 1.3084697627798196,
      "grad_norm": 1.3125,
      "learning_rate": 0.00048327165409284614,
      "loss": 5.0407,
      "mean_token_accuracy": 0.20348008275032042,
      "num_tokens": 28819287.0,
      "step": 15665
    },
    {
      "entropy": 5.314122343063355,
      "epoch": 1.3088874039425327,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004832603880108537,
      "loss": 5.0452,
      "mean_token_accuracy": 0.18996071964502334,
      "num_tokens": 28828907.0,
      "step": 15670
    },
    {
      "entropy": 5.23112998008728,
      "epoch": 1.3093050451052455,
      "grad_norm": 1.28125,
      "learning_rate": 0.00048324911828301256,
      "loss": 4.989,
      "mean_token_accuracy": 0.2004304990172386,
      "num_tokens": 28839087.0,
      "step": 15675
    },
    {
      "entropy": 5.258652400970459,
      "epoch": 1.3097226862679585,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004832378449095199,
      "loss": 5.0634,
      "mean_token_accuracy": 0.20032372772693635,
      "num_tokens": 28848956.0,
      "step": 15680
    },
    {
      "entropy": 5.1796587944030765,
      "epoch": 1.3101403274306715,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004832265678905731,
      "loss": 4.9592,
      "mean_token_accuracy": 0.19994694590568543,
      "num_tokens": 28858105.0,
      "step": 15685
    },
    {
      "entropy": 5.232590866088867,
      "epoch": 1.3105579685933846,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004832152872263697,
      "loss": 5.0836,
      "mean_token_accuracy": 0.19334406703710555,
      "num_tokens": 28867539.0,
      "step": 15690
    },
    {
      "entropy": 5.171653699874878,
      "epoch": 1.3109756097560976,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004832040029171069,
      "loss": 4.9049,
      "mean_token_accuracy": 0.20850061029195785,
      "num_tokens": 28876536.0,
      "step": 15695
    },
    {
      "entropy": 5.226677751541137,
      "epoch": 1.3113932509188104,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00048319271496298255,
      "loss": 5.0222,
      "mean_token_accuracy": 0.192146597802639,
      "num_tokens": 28885354.0,
      "step": 15700
    },
    {
      "entropy": 5.211440849304199,
      "epoch": 1.3118108920815237,
      "grad_norm": 1.296875,
      "learning_rate": 0.00048318142336419413,
      "loss": 4.9422,
      "mean_token_accuracy": 0.20094602257013322,
      "num_tokens": 28894181.0,
      "step": 15705
    },
    {
      "entropy": 5.149627733230591,
      "epoch": 1.3122285332442365,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048317012812093926,
      "loss": 4.925,
      "mean_token_accuracy": 0.2090037554502487,
      "num_tokens": 28903803.0,
      "step": 15710
    },
    {
      "entropy": 5.1720600605010985,
      "epoch": 1.3126461744069495,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004831588292334157,
      "loss": 5.0223,
      "mean_token_accuracy": 0.20072977095842362,
      "num_tokens": 28913192.0,
      "step": 15715
    },
    {
      "entropy": 5.181788206100464,
      "epoch": 1.3130638155696626,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004831475267018213,
      "loss": 4.9326,
      "mean_token_accuracy": 0.20945411771535874,
      "num_tokens": 28922413.0,
      "step": 15720
    },
    {
      "entropy": 5.222289609909057,
      "epoch": 1.3134814567323756,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004831362205263538,
      "loss": 4.9929,
      "mean_token_accuracy": 0.1993620440363884,
      "num_tokens": 28933589.0,
      "step": 15725
    },
    {
      "entropy": 5.196998834609985,
      "epoch": 1.3138990978950886,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004831249107072114,
      "loss": 4.974,
      "mean_token_accuracy": 0.20661796480417252,
      "num_tokens": 28942458.0,
      "step": 15730
    },
    {
      "entropy": 5.220378875732422,
      "epoch": 1.3143167390578014,
      "grad_norm": 1.296875,
      "learning_rate": 0.00048311359724459175,
      "loss": 4.9399,
      "mean_token_accuracy": 0.2098463237285614,
      "num_tokens": 28951834.0,
      "step": 15735
    },
    {
      "entropy": 5.186702585220337,
      "epoch": 1.3147343802205145,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00048310228013869323,
      "loss": 4.9715,
      "mean_token_accuracy": 0.2060798078775406,
      "num_tokens": 28960885.0,
      "step": 15740
    },
    {
      "entropy": 5.19487829208374,
      "epoch": 1.3151520213832275,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004830909593897137,
      "loss": 4.962,
      "mean_token_accuracy": 0.20397440046072007,
      "num_tokens": 28970922.0,
      "step": 15745
    },
    {
      "entropy": 5.241539430618286,
      "epoch": 1.3155696625459405,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004830796349978515,
      "loss": 5.0627,
      "mean_token_accuracy": 0.19841707348823548,
      "num_tokens": 28980873.0,
      "step": 15750
    },
    {
      "entropy": 5.227388858795166,
      "epoch": 1.3159873037086536,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004830683069633048,
      "loss": 5.0379,
      "mean_token_accuracy": 0.19285156726837158,
      "num_tokens": 28989876.0,
      "step": 15755
    },
    {
      "entropy": 5.191253709793091,
      "epoch": 1.3164049448713664,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004830569752862719,
      "loss": 4.9716,
      "mean_token_accuracy": 0.2023260086774826,
      "num_tokens": 28999480.0,
      "step": 15760
    },
    {
      "entropy": 5.183311939239502,
      "epoch": 1.3168225860340796,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00048304563996695127,
      "loss": 5.061,
      "mean_token_accuracy": 0.19626500457525253,
      "num_tokens": 29009855.0,
      "step": 15765
    },
    {
      "entropy": 5.21477255821228,
      "epoch": 1.3172402271967925,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048303430100554126,
      "loss": 4.875,
      "mean_token_accuracy": 0.21433141380548476,
      "num_tokens": 29018506.0,
      "step": 15770
    },
    {
      "entropy": 5.213317394256592,
      "epoch": 1.3176578683595055,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004830229584022404,
      "loss": 4.9501,
      "mean_token_accuracy": 0.20033750534057618,
      "num_tokens": 29027560.0,
      "step": 15775
    },
    {
      "entropy": 5.0726244926452635,
      "epoch": 1.3180755095222185,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004830116121572472,
      "loss": 4.8617,
      "mean_token_accuracy": 0.20473495423793792,
      "num_tokens": 29036038.0,
      "step": 15780
    },
    {
      "entropy": 5.156145191192627,
      "epoch": 1.3184931506849316,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004830002622707604,
      "loss": 4.9343,
      "mean_token_accuracy": 0.20391768515110015,
      "num_tokens": 29046074.0,
      "step": 15785
    },
    {
      "entropy": 5.161706972122192,
      "epoch": 1.3189107918476446,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048298890874297866,
      "loss": 4.9211,
      "mean_token_accuracy": 0.2007024183869362,
      "num_tokens": 29055750.0,
      "step": 15790
    },
    {
      "entropy": 5.138962888717652,
      "epoch": 1.3193284330103574,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004829775515741007,
      "loss": 4.9113,
      "mean_token_accuracy": 0.20102560371160508,
      "num_tokens": 29065012.0,
      "step": 15795
    },
    {
      "entropy": 5.20994119644165,
      "epoch": 1.3197460741730704,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004829661907643253,
      "loss": 5.126,
      "mean_token_accuracy": 0.18616862148046492,
      "num_tokens": 29074780.0,
      "step": 15800
    },
    {
      "entropy": 5.2790015697479244,
      "epoch": 1.3201637153357835,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004829548263138515,
      "loss": 5.0847,
      "mean_token_accuracy": 0.19864239245653154,
      "num_tokens": 29083321.0,
      "step": 15805
    },
    {
      "entropy": 5.17736177444458,
      "epoch": 1.3205813564984965,
      "grad_norm": 1.5,
      "learning_rate": 0.0004829434582228781,
      "loss": 4.9496,
      "mean_token_accuracy": 0.1956817165017128,
      "num_tokens": 29092110.0,
      "step": 15810
    },
    {
      "entropy": 5.109908771514893,
      "epoch": 1.3209989976612095,
      "grad_norm": 1.25,
      "learning_rate": 0.0004829320864916042,
      "loss": 4.8797,
      "mean_token_accuracy": 0.20724192261695862,
      "num_tokens": 29100451.0,
      "step": 15815
    },
    {
      "entropy": 5.253112125396728,
      "epoch": 1.3214166388239224,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00048292071112022886,
      "loss": 4.9771,
      "mean_token_accuracy": 0.19913852959871292,
      "num_tokens": 29110468.0,
      "step": 15820
    },
    {
      "entropy": 5.229425048828125,
      "epoch": 1.3218342799866356,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004829093321089511,
      "loss": 5.0245,
      "mean_token_accuracy": 0.20576322674751282,
      "num_tokens": 29119660.0,
      "step": 15825
    },
    {
      "entropy": 5.265090036392212,
      "epoch": 1.3222519211493484,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004828979494579703,
      "loss": 5.0586,
      "mean_token_accuracy": 0.19453611075878144,
      "num_tokens": 29129145.0,
      "step": 15830
    },
    {
      "entropy": 5.251547908782959,
      "epoch": 1.3226695623120615,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00048288656316748554,
      "loss": 5.0471,
      "mean_token_accuracy": 0.20546285808086395,
      "num_tokens": 29137980.0,
      "step": 15835
    },
    {
      "entropy": 5.1736150741577145,
      "epoch": 1.3230872034747745,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00048287517323769634,
      "loss": 4.9089,
      "mean_token_accuracy": 0.21069984436035155,
      "num_tokens": 29147535.0,
      "step": 15840
    },
    {
      "entropy": 5.067537498474121,
      "epoch": 1.3235048446374875,
      "grad_norm": 1.21875,
      "learning_rate": 0.00048286377966880206,
      "loss": 4.9012,
      "mean_token_accuracy": 0.2097053825855255,
      "num_tokens": 29157162.0,
      "step": 15845
    },
    {
      "entropy": 5.158668088912964,
      "epoch": 1.3239224858002006,
      "grad_norm": 1.1875,
      "learning_rate": 0.00048285238246100203,
      "loss": 4.9334,
      "mean_token_accuracy": 0.20378487557172775,
      "num_tokens": 29166542.0,
      "step": 15850
    },
    {
      "entropy": 5.215177345275879,
      "epoch": 1.3243401269629134,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004828409816144959,
      "loss": 4.9922,
      "mean_token_accuracy": 0.20691298246383666,
      "num_tokens": 29176040.0,
      "step": 15855
    },
    {
      "entropy": 5.271614742279053,
      "epoch": 1.3247577681256264,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004828295771294831,
      "loss": 5.0715,
      "mean_token_accuracy": 0.19815859347581863,
      "num_tokens": 29184730.0,
      "step": 15860
    },
    {
      "entropy": 5.209841585159301,
      "epoch": 1.3251754092883394,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00048281816900616354,
      "loss": 4.9554,
      "mean_token_accuracy": 0.2052948534488678,
      "num_tokens": 29193671.0,
      "step": 15865
    },
    {
      "entropy": 5.185563039779663,
      "epoch": 1.3255930504510525,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004828067572447367,
      "loss": 4.9721,
      "mean_token_accuracy": 0.20800332129001617,
      "num_tokens": 29203094.0,
      "step": 15870
    },
    {
      "entropy": 5.108750915527343,
      "epoch": 1.3260106916137655,
      "grad_norm": 1.53125,
      "learning_rate": 0.00048279534184540246,
      "loss": 4.9604,
      "mean_token_accuracy": 0.209081295132637,
      "num_tokens": 29211678.0,
      "step": 15875
    },
    {
      "entropy": 5.26384425163269,
      "epoch": 1.3264283327764783,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004827839228083606,
      "loss": 5.1041,
      "mean_token_accuracy": 0.19474883526563644,
      "num_tokens": 29220993.0,
      "step": 15880
    },
    {
      "entropy": 5.266596794128418,
      "epoch": 1.3268459739391916,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004827725001338111,
      "loss": 4.9514,
      "mean_token_accuracy": 0.20217220187187196,
      "num_tokens": 29230006.0,
      "step": 15885
    },
    {
      "entropy": 5.189681529998779,
      "epoch": 1.3272636151019044,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048276107382195393,
      "loss": 4.9127,
      "mean_token_accuracy": 0.2130768045783043,
      "num_tokens": 29238906.0,
      "step": 15890
    },
    {
      "entropy": 5.217269515991211,
      "epoch": 1.3276812562646174,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00048274964387298903,
      "loss": 5.0319,
      "mean_token_accuracy": 0.20759542435407638,
      "num_tokens": 29248383.0,
      "step": 15895
    },
    {
      "entropy": 5.13569450378418,
      "epoch": 1.3280988974273304,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004827382102871166,
      "loss": 4.9309,
      "mean_token_accuracy": 0.20457785278558732,
      "num_tokens": 29257999.0,
      "step": 15900
    },
    {
      "entropy": 5.162373828887939,
      "epoch": 1.3285165385900435,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00048272677306453665,
      "loss": 4.9078,
      "mean_token_accuracy": 0.21078399568796158,
      "num_tokens": 29267143.0,
      "step": 15905
    },
    {
      "entropy": 5.221262407302857,
      "epoch": 1.3289341797527565,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00048271533220544955,
      "loss": 5.0569,
      "mean_token_accuracy": 0.19617085605859758,
      "num_tokens": 29276730.0,
      "step": 15910
    },
    {
      "entropy": 5.263863897323608,
      "epoch": 1.3293518209154693,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048270388771005555,
      "loss": 5.0446,
      "mean_token_accuracy": 0.19653396755456926,
      "num_tokens": 29285480.0,
      "step": 15915
    },
    {
      "entropy": 5.235967969894409,
      "epoch": 1.3297694620781824,
      "grad_norm": 1.34375,
      "learning_rate": 0.00048269243957855493,
      "loss": 5.0584,
      "mean_token_accuracy": 0.19528211057186126,
      "num_tokens": 29294927.0,
      "step": 15920
    },
    {
      "entropy": 5.202575540542602,
      "epoch": 1.3301871032408954,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004826809878111481,
      "loss": 4.9681,
      "mean_token_accuracy": 0.20618521124124528,
      "num_tokens": 29304567.0,
      "step": 15925
    },
    {
      "entropy": 5.118433666229248,
      "epoch": 1.3306047444036084,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00048266953240803576,
      "loss": 4.9181,
      "mean_token_accuracy": 0.2085767239332199,
      "num_tokens": 29313302.0,
      "step": 15930
    },
    {
      "entropy": 5.175677919387818,
      "epoch": 1.3310223855663215,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004826580733694182,
      "loss": 4.9344,
      "mean_token_accuracy": 0.20540064573287964,
      "num_tokens": 29322195.0,
      "step": 15935
    },
    {
      "entropy": 5.152914237976074,
      "epoch": 1.3314400267290343,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004826466106954961,
      "loss": 5.0513,
      "mean_token_accuracy": 0.19753078520298004,
      "num_tokens": 29331416.0,
      "step": 15940
    },
    {
      "entropy": 5.206453800201416,
      "epoch": 1.3318576678917475,
      "grad_norm": 2.0625,
      "learning_rate": 0.00048263514438647015,
      "loss": 4.953,
      "mean_token_accuracy": 0.21008970439434052,
      "num_tokens": 29340497.0,
      "step": 15945
    },
    {
      "entropy": 5.113764524459839,
      "epoch": 1.3322753090544603,
      "grad_norm": 1.328125,
      "learning_rate": 0.000482623674442541,
      "loss": 4.927,
      "mean_token_accuracy": 0.21069134622812272,
      "num_tokens": 29349306.0,
      "step": 15950
    },
    {
      "entropy": 5.1907432079315186,
      "epoch": 1.3326929502171734,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004826122008639096,
      "loss": 4.962,
      "mean_token_accuracy": 0.20473988950252534,
      "num_tokens": 29358452.0,
      "step": 15955
    },
    {
      "entropy": 5.201171636581421,
      "epoch": 1.3331105913798864,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00048260072365077675,
      "loss": 5.0315,
      "mean_token_accuracy": 0.19333953112363816,
      "num_tokens": 29367640.0,
      "step": 15960
    },
    {
      "entropy": 5.230400323867798,
      "epoch": 1.3335282325425994,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004825892428033432,
      "loss": 4.9495,
      "mean_token_accuracy": 0.19803310334682464,
      "num_tokens": 29377352.0,
      "step": 15965
    },
    {
      "entropy": 5.1510223865509035,
      "epoch": 1.3339458737053125,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004825777583218102,
      "loss": 4.965,
      "mean_token_accuracy": 0.2007182851433754,
      "num_tokens": 29386802.0,
      "step": 15970
    },
    {
      "entropy": 5.053259181976318,
      "epoch": 1.3343635148680253,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004825662702063787,
      "loss": 4.8339,
      "mean_token_accuracy": 0.2113087072968483,
      "num_tokens": 29395078.0,
      "step": 15975
    },
    {
      "entropy": 5.197069358825684,
      "epoch": 1.3347811560307383,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004825547784572497,
      "loss": 4.9253,
      "mean_token_accuracy": 0.20783321112394332,
      "num_tokens": 29403837.0,
      "step": 15980
    },
    {
      "entropy": 5.179861307144165,
      "epoch": 1.3351987971934514,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004825432830746245,
      "loss": 4.9632,
      "mean_token_accuracy": 0.20326367169618606,
      "num_tokens": 29412222.0,
      "step": 15985
    },
    {
      "entropy": 5.141996431350708,
      "epoch": 1.3356164383561644,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004825317840587044,
      "loss": 4.8817,
      "mean_token_accuracy": 0.21216470301151275,
      "num_tokens": 29421632.0,
      "step": 15990
    },
    {
      "entropy": 5.17267336845398,
      "epoch": 1.3360340795188774,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004825202814096905,
      "loss": 5.0325,
      "mean_token_accuracy": 0.2050176292657852,
      "num_tokens": 29430768.0,
      "step": 15995
    },
    {
      "entropy": 5.342801427841186,
      "epoch": 1.3364517206815905,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004825087751277843,
      "loss": 5.0681,
      "mean_token_accuracy": 0.19845687001943588,
      "num_tokens": 29440280.0,
      "step": 16000
    },
    {
      "entropy": 5.313129425048828,
      "epoch": 1.3368693618443035,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00048249726521318716,
      "loss": 4.9732,
      "mean_token_accuracy": 0.20272117257118225,
      "num_tokens": 29449850.0,
      "step": 16005
    },
    {
      "entropy": 5.165170431137085,
      "epoch": 1.3372870030070163,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004824857516661008,
      "loss": 4.9204,
      "mean_token_accuracy": 0.21130103170871734,
      "num_tokens": 29459233.0,
      "step": 16010
    },
    {
      "entropy": 5.207938480377197,
      "epoch": 1.3377046441697293,
      "grad_norm": 1.15625,
      "learning_rate": 0.00048247423448672647,
      "loss": 5.043,
      "mean_token_accuracy": 0.1969639018177986,
      "num_tokens": 29469351.0,
      "step": 16015
    },
    {
      "entropy": 5.1647687435150145,
      "epoch": 1.3381222853324424,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004824627136752659,
      "loss": 5.0192,
      "mean_token_accuracy": 0.20145349949598312,
      "num_tokens": 29477653.0,
      "step": 16020
    },
    {
      "entropy": 5.204736948013306,
      "epoch": 1.3385399264951554,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004824511892319209,
      "loss": 4.96,
      "mean_token_accuracy": 0.207114377617836,
      "num_tokens": 29486944.0,
      "step": 16025
    },
    {
      "entropy": 5.121522235870361,
      "epoch": 1.3389575676578684,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00048243966115689305,
      "loss": 4.9045,
      "mean_token_accuracy": 0.21261273324489594,
      "num_tokens": 29495791.0,
      "step": 16030
    },
    {
      "entropy": 5.2959826469421385,
      "epoch": 1.3393752088205813,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004824281294503843,
      "loss": 5.1707,
      "mean_token_accuracy": 0.19179708361625672,
      "num_tokens": 29505403.0,
      "step": 16035
    },
    {
      "entropy": 5.1597062110900875,
      "epoch": 1.3397928499832943,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004824165941125964,
      "loss": 4.847,
      "mean_token_accuracy": 0.20796542316675187,
      "num_tokens": 29515310.0,
      "step": 16040
    },
    {
      "entropy": 5.2043726444244385,
      "epoch": 1.3402104911460073,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004824050551437315,
      "loss": 5.0301,
      "mean_token_accuracy": 0.20401650965213775,
      "num_tokens": 29524824.0,
      "step": 16045
    },
    {
      "entropy": 5.146274566650391,
      "epoch": 1.3406281323087204,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004823935125439912,
      "loss": 4.927,
      "mean_token_accuracy": 0.2029123857617378,
      "num_tokens": 29533667.0,
      "step": 16050
    },
    {
      "entropy": 5.218888187408448,
      "epoch": 1.3410457734714334,
      "grad_norm": 1.3125,
      "learning_rate": 0.000482381966313578,
      "loss": 5.056,
      "mean_token_accuracy": 0.1995869517326355,
      "num_tokens": 29541745.0,
      "step": 16055
    },
    {
      "entropy": 5.158271980285645,
      "epoch": 1.3414634146341464,
      "grad_norm": 1.15625,
      "learning_rate": 0.00048237041645269387,
      "loss": 4.8857,
      "mean_token_accuracy": 0.2081284150481224,
      "num_tokens": 29551004.0,
      "step": 16060
    },
    {
      "entropy": 5.0771246433258055,
      "epoch": 1.3418810557968595,
      "grad_norm": 1.25,
      "learning_rate": 0.0004823588629615409,
      "loss": 4.9443,
      "mean_token_accuracy": 0.20688118040561676,
      "num_tokens": 29561061.0,
      "step": 16065
    },
    {
      "entropy": 5.078748321533203,
      "epoch": 1.3422986969595723,
      "grad_norm": 1.28125,
      "learning_rate": 0.00048234730584032143,
      "loss": 4.7659,
      "mean_token_accuracy": 0.22053611129522324,
      "num_tokens": 29569689.0,
      "step": 16070
    },
    {
      "entropy": 5.176053667068482,
      "epoch": 1.3427163381222853,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004823357450892379,
      "loss": 5.0266,
      "mean_token_accuracy": 0.19800310879945754,
      "num_tokens": 29579576.0,
      "step": 16075
    },
    {
      "entropy": 5.14707202911377,
      "epoch": 1.3431339792849983,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004823241807084925,
      "loss": 4.8876,
      "mean_token_accuracy": 0.21138682812452317,
      "num_tokens": 29588531.0,
      "step": 16080
    },
    {
      "entropy": 5.230626821517944,
      "epoch": 1.3435516204477114,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00048231261269828776,
      "loss": 4.9756,
      "mean_token_accuracy": 0.20204796642065048,
      "num_tokens": 29596838.0,
      "step": 16085
    },
    {
      "entropy": 5.1817316055297855,
      "epoch": 1.3439692616104244,
      "grad_norm": 1.34375,
      "learning_rate": 0.00048230104105882624,
      "loss": 4.9613,
      "mean_token_accuracy": 0.2058747723698616,
      "num_tokens": 29606769.0,
      "step": 16090
    },
    {
      "entropy": 5.215613508224488,
      "epoch": 1.3443869027731372,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004822894657903104,
      "loss": 5.0511,
      "mean_token_accuracy": 0.1956161752343178,
      "num_tokens": 29616994.0,
      "step": 16095
    },
    {
      "entropy": 5.092758417129517,
      "epoch": 1.3448045439358502,
      "grad_norm": 1.21875,
      "learning_rate": 0.000482277886892943,
      "loss": 4.9189,
      "mean_token_accuracy": 0.21360840499401093,
      "num_tokens": 29626865.0,
      "step": 16100
    },
    {
      "entropy": 5.169208669662476,
      "epoch": 1.3452221850985633,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004822663043669267,
      "loss": 4.8555,
      "mean_token_accuracy": 0.2109660342335701,
      "num_tokens": 29636439.0,
      "step": 16105
    },
    {
      "entropy": 5.162726306915284,
      "epoch": 1.3456398262612763,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048225471821246416,
      "loss": 4.8564,
      "mean_token_accuracy": 0.2124568060040474,
      "num_tokens": 29645019.0,
      "step": 16110
    },
    {
      "entropy": 5.166385984420776,
      "epoch": 1.3460574674239894,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004822431284297584,
      "loss": 5.1082,
      "mean_token_accuracy": 0.2033352419734001,
      "num_tokens": 29654660.0,
      "step": 16115
    },
    {
      "entropy": 5.186561155319214,
      "epoch": 1.3464751085867024,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004822315350190122,
      "loss": 5.0088,
      "mean_token_accuracy": 0.1974615201354027,
      "num_tokens": 29662880.0,
      "step": 16120
    },
    {
      "entropy": 5.261029863357544,
      "epoch": 1.3468927497494154,
      "grad_norm": 1.21875,
      "learning_rate": 0.00048221993798042856,
      "loss": 5.0091,
      "mean_token_accuracy": 0.19690718054771422,
      "num_tokens": 29672336.0,
      "step": 16125
    },
    {
      "entropy": 5.1601417541503904,
      "epoch": 1.3473103909121282,
      "grad_norm": 1.40625,
      "learning_rate": 0.00048220833731421045,
      "loss": 4.9609,
      "mean_token_accuracy": 0.19782706648111342,
      "num_tokens": 29681495.0,
      "step": 16130
    },
    {
      "entropy": 5.204880046844482,
      "epoch": 1.3477280320748413,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004821967330205609,
      "loss": 4.9412,
      "mean_token_accuracy": 0.20552052110433577,
      "num_tokens": 29690214.0,
      "step": 16135
    },
    {
      "entropy": 5.191518640518188,
      "epoch": 1.3481456732375543,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004821851250996832,
      "loss": 5.0001,
      "mean_token_accuracy": 0.19943521320819854,
      "num_tokens": 29699309.0,
      "step": 16140
    },
    {
      "entropy": 5.137046146392822,
      "epoch": 1.3485633144002673,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004821735135517805,
      "loss": 4.9149,
      "mean_token_accuracy": 0.20391877591609955,
      "num_tokens": 29708741.0,
      "step": 16145
    },
    {
      "entropy": 5.141409873962402,
      "epoch": 1.3489809555629804,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004821618983770561,
      "loss": 5.0357,
      "mean_token_accuracy": 0.1993741974234581,
      "num_tokens": 29717616.0,
      "step": 16150
    },
    {
      "entropy": 5.212161636352539,
      "epoch": 1.3493985967256932,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00048215027957571326,
      "loss": 4.9204,
      "mean_token_accuracy": 0.20475374907255173,
      "num_tokens": 29726601.0,
      "step": 16155
    },
    {
      "entropy": 5.219625997543335,
      "epoch": 1.3498162378884062,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004821386571479555,
      "loss": 4.9994,
      "mean_token_accuracy": 0.1998400241136551,
      "num_tokens": 29736930.0,
      "step": 16160
    },
    {
      "entropy": 5.109027576446533,
      "epoch": 1.3502338790511192,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004821270310939861,
      "loss": 4.8672,
      "mean_token_accuracy": 0.2095703050494194,
      "num_tokens": 29745980.0,
      "step": 16165
    },
    {
      "entropy": 5.163994359970093,
      "epoch": 1.3506515202138323,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004821154014140088,
      "loss": 4.9663,
      "mean_token_accuracy": 0.20562876164913177,
      "num_tokens": 29755701.0,
      "step": 16170
    },
    {
      "entropy": 5.1276243209838865,
      "epoch": 1.3510691613765453,
      "grad_norm": 1.25,
      "learning_rate": 0.0004821037681082271,
      "loss": 4.8544,
      "mean_token_accuracy": 0.21383642852306367,
      "num_tokens": 29763503.0,
      "step": 16175
    },
    {
      "entropy": 5.076704502105713,
      "epoch": 1.3514868025392583,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004820921311768446,
      "loss": 4.8264,
      "mean_token_accuracy": 0.20840473026037215,
      "num_tokens": 29771667.0,
      "step": 16180
    },
    {
      "entropy": 5.190806770324707,
      "epoch": 1.3519044437019714,
      "grad_norm": 1.25,
      "learning_rate": 0.00048208049062006505,
      "loss": 5.021,
      "mean_token_accuracy": 0.19785860031843186,
      "num_tokens": 29780836.0,
      "step": 16185
    },
    {
      "entropy": 5.175419569015503,
      "epoch": 1.3523220848646842,
      "grad_norm": 1.34375,
      "learning_rate": 0.00048206884643809224,
      "loss": 4.9468,
      "mean_token_accuracy": 0.20445718765258789,
      "num_tokens": 29790397.0,
      "step": 16190
    },
    {
      "entropy": 5.173952579498291,
      "epoch": 1.3527397260273972,
      "grad_norm": 1.375,
      "learning_rate": 0.00048205719863113014,
      "loss": 5.0047,
      "mean_token_accuracy": 0.20499362498521806,
      "num_tokens": 29799227.0,
      "step": 16195
    },
    {
      "entropy": 5.202955341339111,
      "epoch": 1.3531573671901103,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004820455471993824,
      "loss": 4.9964,
      "mean_token_accuracy": 0.19886355847120285,
      "num_tokens": 29808182.0,
      "step": 16200
    },
    {
      "entropy": 5.224887180328369,
      "epoch": 1.3535750083528233,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00048203389214305326,
      "loss": 5.0238,
      "mean_token_accuracy": 0.19891393929719925,
      "num_tokens": 29817843.0,
      "step": 16205
    },
    {
      "entropy": 5.192651081085205,
      "epoch": 1.3539926495155363,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004820222334623465,
      "loss": 4.9439,
      "mean_token_accuracy": 0.20748689025640488,
      "num_tokens": 29827082.0,
      "step": 16210
    },
    {
      "entropy": 5.205419063568115,
      "epoch": 1.3544102906782491,
      "grad_norm": 1.453125,
      "learning_rate": 0.00048201057115746643,
      "loss": 4.9288,
      "mean_token_accuracy": 0.20176009833812714,
      "num_tokens": 29836176.0,
      "step": 16215
    },
    {
      "entropy": 5.116638231277466,
      "epoch": 1.3548279318409622,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004819989052286172,
      "loss": 4.9008,
      "mean_token_accuracy": 0.20876089930534364,
      "num_tokens": 29845252.0,
      "step": 16220
    },
    {
      "entropy": 5.147748136520386,
      "epoch": 1.3552455730036752,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004819872356760028,
      "loss": 4.9408,
      "mean_token_accuracy": 0.20843307673931122,
      "num_tokens": 29854115.0,
      "step": 16225
    },
    {
      "entropy": 5.180002307891845,
      "epoch": 1.3556632141663882,
      "grad_norm": 1.359375,
      "learning_rate": 0.00048197556249982774,
      "loss": 4.9621,
      "mean_token_accuracy": 0.21044610142707826,
      "num_tokens": 29861918.0,
      "step": 16230
    },
    {
      "entropy": 5.241896772384644,
      "epoch": 1.3560808553291013,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004819638857002965,
      "loss": 4.9713,
      "mean_token_accuracy": 0.20431248992681503,
      "num_tokens": 29871421.0,
      "step": 16235
    },
    {
      "entropy": 5.175195980072021,
      "epoch": 1.3564984964918143,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048195220527761306,
      "loss": 4.8798,
      "mean_token_accuracy": 0.21112172156572342,
      "num_tokens": 29880328.0,
      "step": 16240
    },
    {
      "entropy": 5.231692266464234,
      "epoch": 1.3569161376545273,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00048194052123198236,
      "loss": 5.084,
      "mean_token_accuracy": 0.19766909182071685,
      "num_tokens": 29889777.0,
      "step": 16245
    },
    {
      "entropy": 5.1463099956512455,
      "epoch": 1.3573337788172402,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00048192883356360867,
      "loss": 4.9202,
      "mean_token_accuracy": 0.20488062351942063,
      "num_tokens": 29898638.0,
      "step": 16250
    },
    {
      "entropy": 5.1478235721588135,
      "epoch": 1.3577514199799532,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004819171422726966,
      "loss": 4.9435,
      "mean_token_accuracy": 0.20461610406637193,
      "num_tokens": 29907856.0,
      "step": 16255
    },
    {
      "entropy": 5.180762243270874,
      "epoch": 1.3581690611426662,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00048190544735945094,
      "loss": 5.0215,
      "mean_token_accuracy": 0.19685481637716293,
      "num_tokens": 29916992.0,
      "step": 16260
    },
    {
      "entropy": 5.213744831085205,
      "epoch": 1.3585867023053793,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004818937488240764,
      "loss": 4.9396,
      "mean_token_accuracy": 0.20635010153055192,
      "num_tokens": 29926323.0,
      "step": 16265
    },
    {
      "entropy": 5.205189323425293,
      "epoch": 1.3590043434680923,
      "grad_norm": 1.3125,
      "learning_rate": 0.00048188204666677775,
      "loss": 5.0015,
      "mean_token_accuracy": 0.19704319834709166,
      "num_tokens": 29935386.0,
      "step": 16270
    },
    {
      "entropy": 5.20824122428894,
      "epoch": 1.359421984630805,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004818703408877598,
      "loss": 4.9897,
      "mean_token_accuracy": 0.20356898307800292,
      "num_tokens": 29944741.0,
      "step": 16275
    },
    {
      "entropy": 5.181627798080444,
      "epoch": 1.3598396257935181,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004818586314872276,
      "loss": 4.8944,
      "mean_token_accuracy": 0.20571573376655578,
      "num_tokens": 29952407.0,
      "step": 16280
    },
    {
      "entropy": 5.143144321441651,
      "epoch": 1.3602572669562312,
      "grad_norm": 1.421875,
      "learning_rate": 0.00048184691846538596,
      "loss": 4.9246,
      "mean_token_accuracy": 0.20104315578937532,
      "num_tokens": 29962315.0,
      "step": 16285
    },
    {
      "entropy": 5.173208236694336,
      "epoch": 1.3606749081189442,
      "grad_norm": 1.34375,
      "learning_rate": 0.00048183520182244006,
      "loss": 5.0001,
      "mean_token_accuracy": 0.20647431313991546,
      "num_tokens": 29972070.0,
      "step": 16290
    },
    {
      "entropy": 5.171013927459716,
      "epoch": 1.3610925492816572,
      "grad_norm": 1.28125,
      "learning_rate": 0.000481823481558595,
      "loss": 4.9044,
      "mean_token_accuracy": 0.2108625665307045,
      "num_tokens": 29980416.0,
      "step": 16295
    },
    {
      "entropy": 5.2695259094238285,
      "epoch": 1.3615101904443703,
      "grad_norm": 1.453125,
      "learning_rate": 0.00048181175767405594,
      "loss": 4.9829,
      "mean_token_accuracy": 0.20483639538288118,
      "num_tokens": 29989436.0,
      "step": 16300
    },
    {
      "entropy": 5.172491264343262,
      "epoch": 1.3619278316070833,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004818000301690282,
      "loss": 4.8932,
      "mean_token_accuracy": 0.2066403642296791,
      "num_tokens": 29998295.0,
      "step": 16305
    },
    {
      "entropy": 5.090137195587158,
      "epoch": 1.3623454727697961,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004817882990437169,
      "loss": 4.9445,
      "mean_token_accuracy": 0.2008734554052353,
      "num_tokens": 30007892.0,
      "step": 16310
    },
    {
      "entropy": 5.147916650772094,
      "epoch": 1.3627631139325092,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00048177656429832763,
      "loss": 4.9005,
      "mean_token_accuracy": 0.21236618757247924,
      "num_tokens": 30016477.0,
      "step": 16315
    },
    {
      "entropy": 5.060186195373535,
      "epoch": 1.3631807550952222,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00048176482593306557,
      "loss": 4.9082,
      "mean_token_accuracy": 0.20835582613945008,
      "num_tokens": 30025693.0,
      "step": 16320
    },
    {
      "entropy": 5.235565042495727,
      "epoch": 1.3635983962579352,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00048175308394813644,
      "loss": 4.9247,
      "mean_token_accuracy": 0.2023969680070877,
      "num_tokens": 30034552.0,
      "step": 16325
    },
    {
      "entropy": 5.139854669570923,
      "epoch": 1.3640160374206483,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004817413383437456,
      "loss": 4.9235,
      "mean_token_accuracy": 0.21011175364255905,
      "num_tokens": 30043584.0,
      "step": 16330
    },
    {
      "entropy": 5.220077323913574,
      "epoch": 1.364433678583361,
      "grad_norm": 1.296875,
      "learning_rate": 0.00048172958912009886,
      "loss": 5.0504,
      "mean_token_accuracy": 0.19872331023216247,
      "num_tokens": 30053189.0,
      "step": 16335
    },
    {
      "entropy": 5.116402673721313,
      "epoch": 1.364851319746074,
      "grad_norm": 1.28125,
      "learning_rate": 0.00048171783627740176,
      "loss": 4.8964,
      "mean_token_accuracy": 0.2032360091805458,
      "num_tokens": 30062144.0,
      "step": 16340
    },
    {
      "entropy": 5.205732536315918,
      "epoch": 1.3652689609087871,
      "grad_norm": 1.3125,
      "learning_rate": 0.00048170607981586023,
      "loss": 4.9281,
      "mean_token_accuracy": 0.20388443917036056,
      "num_tokens": 30071646.0,
      "step": 16345
    },
    {
      "entropy": 5.240680694580078,
      "epoch": 1.3656866020715002,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004816943197356799,
      "loss": 5.0431,
      "mean_token_accuracy": 0.20513294637203217,
      "num_tokens": 30081322.0,
      "step": 16350
    },
    {
      "entropy": 5.090106439590454,
      "epoch": 1.3661042432342132,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004816825560370667,
      "loss": 4.8999,
      "mean_token_accuracy": 0.21650386601686478,
      "num_tokens": 30090059.0,
      "step": 16355
    },
    {
      "entropy": 5.244268989562988,
      "epoch": 1.3665218843969262,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00048167078872022657,
      "loss": 5.0431,
      "mean_token_accuracy": 0.2039051443338394,
      "num_tokens": 30100924.0,
      "step": 16360
    },
    {
      "entropy": 5.158176612854004,
      "epoch": 1.3669395255596393,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00048165901778536547,
      "loss": 4.9087,
      "mean_token_accuracy": 0.20888210237026214,
      "num_tokens": 30109972.0,
      "step": 16365
    },
    {
      "entropy": 5.191336917877197,
      "epoch": 1.367357166722352,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00048164724323268956,
      "loss": 4.9246,
      "mean_token_accuracy": 0.2013370454311371,
      "num_tokens": 30118705.0,
      "step": 16370
    },
    {
      "entropy": 5.189301919937134,
      "epoch": 1.3677748078850651,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00048163546506240493,
      "loss": 5.0446,
      "mean_token_accuracy": 0.2040234163403511,
      "num_tokens": 30127631.0,
      "step": 16375
    },
    {
      "entropy": 5.125183486938477,
      "epoch": 1.3681924490477781,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004816236832747177,
      "loss": 4.8894,
      "mean_token_accuracy": 0.2127163901925087,
      "num_tokens": 30136127.0,
      "step": 16380
    },
    {
      "entropy": 5.128161525726318,
      "epoch": 1.3686100902104912,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004816118978698342,
      "loss": 4.9191,
      "mean_token_accuracy": 0.20590122789144516,
      "num_tokens": 30145609.0,
      "step": 16385
    },
    {
      "entropy": 5.322847127914429,
      "epoch": 1.3690277313732042,
      "grad_norm": 1.375,
      "learning_rate": 0.0004816001088479608,
      "loss": 5.0562,
      "mean_token_accuracy": 0.19165219515562057,
      "num_tokens": 30155705.0,
      "step": 16390
    },
    {
      "entropy": 5.1907049179077145,
      "epoch": 1.369445372535917,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004815883162093037,
      "loss": 5.006,
      "mean_token_accuracy": 0.19634098261594773,
      "num_tokens": 30165549.0,
      "step": 16395
    },
    {
      "entropy": 5.122199535369873,
      "epoch": 1.36986301369863,
      "grad_norm": 1.421875,
      "learning_rate": 0.00048157651995406954,
      "loss": 4.9564,
      "mean_token_accuracy": 0.2032659351825714,
      "num_tokens": 30174716.0,
      "step": 16400
    },
    {
      "entropy": 5.162789344787598,
      "epoch": 1.370280654861343,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004815647200824647,
      "loss": 4.9097,
      "mean_token_accuracy": 0.20668301731348038,
      "num_tokens": 30184009.0,
      "step": 16405
    },
    {
      "entropy": 5.2835382461547855,
      "epoch": 1.3706982960240561,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00048155291659469587,
      "loss": 5.0023,
      "mean_token_accuracy": 0.19937612265348434,
      "num_tokens": 30193731.0,
      "step": 16410
    },
    {
      "entropy": 5.17013201713562,
      "epoch": 1.3711159371867692,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00048154110949096945,
      "loss": 5.1167,
      "mean_token_accuracy": 0.19380486011505127,
      "num_tokens": 30203643.0,
      "step": 16415
    },
    {
      "entropy": 5.192875242233276,
      "epoch": 1.3715335783494822,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004815292987714924,
      "loss": 4.9483,
      "mean_token_accuracy": 0.2036214292049408,
      "num_tokens": 30212436.0,
      "step": 16420
    },
    {
      "entropy": 5.169115495681763,
      "epoch": 1.3719512195121952,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004815174844364714,
      "loss": 4.899,
      "mean_token_accuracy": 0.20918850898742675,
      "num_tokens": 30221730.0,
      "step": 16425
    },
    {
      "entropy": 5.243129730224609,
      "epoch": 1.372368860674908,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004815056664861133,
      "loss": 4.9659,
      "mean_token_accuracy": 0.20117760002613067,
      "num_tokens": 30230645.0,
      "step": 16430
    },
    {
      "entropy": 5.125395965576172,
      "epoch": 1.372786501837621,
      "grad_norm": 1.375,
      "learning_rate": 0.0004814938449206248,
      "loss": 4.933,
      "mean_token_accuracy": 0.20487329065799714,
      "num_tokens": 30239513.0,
      "step": 16435
    },
    {
      "entropy": 5.112496185302734,
      "epoch": 1.373204143000334,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00048148201974021297,
      "loss": 4.9163,
      "mean_token_accuracy": 0.2073461577296257,
      "num_tokens": 30248920.0,
      "step": 16440
    },
    {
      "entropy": 5.148032474517822,
      "epoch": 1.3736217841630471,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000481470190945085,
      "loss": 4.9137,
      "mean_token_accuracy": 0.20689001828432083,
      "num_tokens": 30257851.0,
      "step": 16445
    },
    {
      "entropy": 5.104607582092285,
      "epoch": 1.3740394253257602,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004814583585354477,
      "loss": 4.9422,
      "mean_token_accuracy": 0.21386310011148452,
      "num_tokens": 30267676.0,
      "step": 16450
    },
    {
      "entropy": 5.093468379974365,
      "epoch": 1.374457066488473,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00048144652251150836,
      "loss": 4.8335,
      "mean_token_accuracy": 0.20858850330114365,
      "num_tokens": 30277744.0,
      "step": 16455
    },
    {
      "entropy": 5.229084539413452,
      "epoch": 1.374874707651186,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000481434682873474,
      "loss": 4.9851,
      "mean_token_accuracy": 0.1943365976214409,
      "num_tokens": 30286578.0,
      "step": 16460
    },
    {
      "entropy": 5.2041984558105465,
      "epoch": 1.375292348813899,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004814228396215522,
      "loss": 4.9765,
      "mean_token_accuracy": 0.20536316186189651,
      "num_tokens": 30295896.0,
      "step": 16465
    },
    {
      "entropy": 5.20637149810791,
      "epoch": 1.375709989976612,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004814109927559501,
      "loss": 5.0412,
      "mean_token_accuracy": 0.19787871837615967,
      "num_tokens": 30305397.0,
      "step": 16470
    },
    {
      "entropy": 5.171839570999145,
      "epoch": 1.3761276311393251,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004813991422768751,
      "loss": 4.9908,
      "mean_token_accuracy": 0.21076744496822358,
      "num_tokens": 30314025.0,
      "step": 16475
    },
    {
      "entropy": 5.169173240661621,
      "epoch": 1.3765452723020382,
      "grad_norm": 1.578125,
      "learning_rate": 0.00048138728818453456,
      "loss": 4.9836,
      "mean_token_accuracy": 0.19945336282253265,
      "num_tokens": 30324384.0,
      "step": 16480
    },
    {
      "entropy": 5.213486766815185,
      "epoch": 1.3769629134647512,
      "grad_norm": 1.1875,
      "learning_rate": 0.00048137543047913636,
      "loss": 4.9956,
      "mean_token_accuracy": 0.20578465312719346,
      "num_tokens": 30333663.0,
      "step": 16485
    },
    {
      "entropy": 5.166088819503784,
      "epoch": 1.377380554627464,
      "grad_norm": 1.28125,
      "learning_rate": 0.00048136356916088764,
      "loss": 4.9519,
      "mean_token_accuracy": 0.1982530653476715,
      "num_tokens": 30343603.0,
      "step": 16490
    },
    {
      "entropy": 5.190098285675049,
      "epoch": 1.377798195790177,
      "grad_norm": 1.296875,
      "learning_rate": 0.00048135170422999624,
      "loss": 5.0175,
      "mean_token_accuracy": 0.20566751956939697,
      "num_tokens": 30352680.0,
      "step": 16495
    },
    {
      "entropy": 5.189872455596924,
      "epoch": 1.37821583695289,
      "grad_norm": 1.25,
      "learning_rate": 0.00048133983568666994,
      "loss": 4.9746,
      "mean_token_accuracy": 0.20408557653427123,
      "num_tokens": 30362597.0,
      "step": 16500
    },
    {
      "entropy": 5.32915096282959,
      "epoch": 1.378633478115603,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004813279635311165,
      "loss": 5.101,
      "mean_token_accuracy": 0.1952545687556267,
      "num_tokens": 30372037.0,
      "step": 16505
    },
    {
      "entropy": 5.216665744781494,
      "epoch": 1.3790511192783161,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004813160877635436,
      "loss": 4.9593,
      "mean_token_accuracy": 0.20490257441997528,
      "num_tokens": 30380867.0,
      "step": 16510
    },
    {
      "entropy": 5.201509618759156,
      "epoch": 1.379468760441029,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00048130420838415926,
      "loss": 5.0309,
      "mean_token_accuracy": 0.1962714523077011,
      "num_tokens": 30391181.0,
      "step": 16515
    },
    {
      "entropy": 5.188954925537109,
      "epoch": 1.379886401603742,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048129232539317147,
      "loss": 5.0198,
      "mean_token_accuracy": 0.20080854892730712,
      "num_tokens": 30400519.0,
      "step": 16520
    },
    {
      "entropy": 5.1658362865448,
      "epoch": 1.380304042766455,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00048128043879078816,
      "loss": 4.9217,
      "mean_token_accuracy": 0.20232172012329103,
      "num_tokens": 30409570.0,
      "step": 16525
    },
    {
      "entropy": 5.238266372680664,
      "epoch": 1.380721683929168,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004812685485772175,
      "loss": 4.9575,
      "mean_token_accuracy": 0.20328852385282517,
      "num_tokens": 30418458.0,
      "step": 16530
    },
    {
      "entropy": 5.205003833770752,
      "epoch": 1.381139325091881,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00048125665475266756,
      "loss": 4.9652,
      "mean_token_accuracy": 0.20482176542282104,
      "num_tokens": 30427933.0,
      "step": 16535
    },
    {
      "entropy": 5.251329326629639,
      "epoch": 1.3815569662545941,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00048124475731734667,
      "loss": 5.0369,
      "mean_token_accuracy": 0.20321532785892488,
      "num_tokens": 30436825.0,
      "step": 16540
    },
    {
      "entropy": 5.174755048751831,
      "epoch": 1.3819746074173072,
      "grad_norm": 1.4453125,
      "learning_rate": 0.000481232856271463,
      "loss": 4.9502,
      "mean_token_accuracy": 0.20051774233579636,
      "num_tokens": 30445367.0,
      "step": 16545
    },
    {
      "entropy": 5.25114221572876,
      "epoch": 1.38239224858002,
      "grad_norm": 1.34375,
      "learning_rate": 0.00048122095161522496,
      "loss": 5.0754,
      "mean_token_accuracy": 0.1975399672985077,
      "num_tokens": 30455236.0,
      "step": 16550
    },
    {
      "entropy": 5.150349092483521,
      "epoch": 1.382809889742733,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004812090433488409,
      "loss": 4.9218,
      "mean_token_accuracy": 0.20656113475561141,
      "num_tokens": 30465003.0,
      "step": 16555
    },
    {
      "entropy": 5.167342042922973,
      "epoch": 1.383227530905446,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048119713147251944,
      "loss": 4.9052,
      "mean_token_accuracy": 0.20629769712686538,
      "num_tokens": 30473686.0,
      "step": 16560
    },
    {
      "entropy": 5.17631049156189,
      "epoch": 1.383645172068159,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048118521598646886,
      "loss": 5.0222,
      "mean_token_accuracy": 0.20142582952976226,
      "num_tokens": 30482684.0,
      "step": 16565
    },
    {
      "entropy": 5.1887775421142575,
      "epoch": 1.384062813230872,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00048117329689089793,
      "loss": 4.8851,
      "mean_token_accuracy": 0.2075921595096588,
      "num_tokens": 30491356.0,
      "step": 16570
    },
    {
      "entropy": 5.19974946975708,
      "epoch": 1.384480454393585,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00048116137418601526,
      "loss": 4.9245,
      "mean_token_accuracy": 0.20036507546901702,
      "num_tokens": 30500829.0,
      "step": 16575
    },
    {
      "entropy": 5.152085542678833,
      "epoch": 1.384898095556298,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00048114944787202955,
      "loss": 5.059,
      "mean_token_accuracy": 0.19590733647346498,
      "num_tokens": 30509759.0,
      "step": 16580
    },
    {
      "entropy": 5.136808776855469,
      "epoch": 1.385315736719011,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00048113751794914974,
      "loss": 4.881,
      "mean_token_accuracy": 0.2108369544148445,
      "num_tokens": 30518638.0,
      "step": 16585
    },
    {
      "entropy": 5.148374032974243,
      "epoch": 1.385733377881724,
      "grad_norm": 1.53125,
      "learning_rate": 0.00048112558441758437,
      "loss": 4.9197,
      "mean_token_accuracy": 0.20694513469934464,
      "num_tokens": 30527973.0,
      "step": 16590
    },
    {
      "entropy": 5.129864025115967,
      "epoch": 1.386151019044437,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004811136472775426,
      "loss": 4.8726,
      "mean_token_accuracy": 0.2095533013343811,
      "num_tokens": 30537120.0,
      "step": 16595
    },
    {
      "entropy": 5.253251266479492,
      "epoch": 1.38656866020715,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004811017065292334,
      "loss": 4.9629,
      "mean_token_accuracy": 0.20243980288505553,
      "num_tokens": 30546372.0,
      "step": 16600
    },
    {
      "entropy": 5.249629306793213,
      "epoch": 1.3869863013698631,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00048108976217286576,
      "loss": 4.9698,
      "mean_token_accuracy": 0.2015757903456688,
      "num_tokens": 30555377.0,
      "step": 16605
    },
    {
      "entropy": 5.087387275695801,
      "epoch": 1.387403942532576,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004810778142086487,
      "loss": 4.9366,
      "mean_token_accuracy": 0.20814029574394227,
      "num_tokens": 30565208.0,
      "step": 16610
    },
    {
      "entropy": 5.12762861251831,
      "epoch": 1.387821583695289,
      "grad_norm": 1.359375,
      "learning_rate": 0.00048106586263679146,
      "loss": 4.9922,
      "mean_token_accuracy": 0.20747557580471038,
      "num_tokens": 30573230.0,
      "step": 16615
    },
    {
      "entropy": 5.194325685501099,
      "epoch": 1.388239224858002,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00048105390745750324,
      "loss": 4.866,
      "mean_token_accuracy": 0.21104021966457367,
      "num_tokens": 30581570.0,
      "step": 16620
    },
    {
      "entropy": 5.1505879878997805,
      "epoch": 1.388656866020715,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00048104194867099336,
      "loss": 4.9462,
      "mean_token_accuracy": 0.20430634915828705,
      "num_tokens": 30591091.0,
      "step": 16625
    },
    {
      "entropy": 5.095681190490723,
      "epoch": 1.389074507183428,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00048102998627747116,
      "loss": 4.9179,
      "mean_token_accuracy": 0.20857519209384917,
      "num_tokens": 30601545.0,
      "step": 16630
    },
    {
      "entropy": 5.232680797576904,
      "epoch": 1.3894921483461409,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000481018020277146,
      "loss": 4.9844,
      "mean_token_accuracy": 0.196837817132473,
      "num_tokens": 30610630.0,
      "step": 16635
    },
    {
      "entropy": 5.135907173156738,
      "epoch": 1.389909789508854,
      "grad_norm": 1.25,
      "learning_rate": 0.00048100605067022753,
      "loss": 4.8863,
      "mean_token_accuracy": 0.20805514752864837,
      "num_tokens": 30620015.0,
      "step": 16640
    },
    {
      "entropy": 5.204732227325439,
      "epoch": 1.390327430671567,
      "grad_norm": 1.296875,
      "learning_rate": 0.00048099407745692504,
      "loss": 4.8891,
      "mean_token_accuracy": 0.21397576630115508,
      "num_tokens": 30628590.0,
      "step": 16645
    },
    {
      "entropy": 5.125464582443238,
      "epoch": 1.39074507183428,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00048098210063744827,
      "loss": 4.9385,
      "mean_token_accuracy": 0.21014172583818436,
      "num_tokens": 30638624.0,
      "step": 16650
    },
    {
      "entropy": 5.186296844482422,
      "epoch": 1.391162712996993,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00048097012021200696,
      "loss": 4.9938,
      "mean_token_accuracy": 0.19819470196962358,
      "num_tokens": 30648247.0,
      "step": 16655
    },
    {
      "entropy": 5.271215152740479,
      "epoch": 1.391580354159706,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004809581361808108,
      "loss": 5.0935,
      "mean_token_accuracy": 0.20142492055892944,
      "num_tokens": 30659045.0,
      "step": 16660
    },
    {
      "entropy": 5.2340654850006105,
      "epoch": 1.391997995322419,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004809461485440694,
      "loss": 4.9679,
      "mean_token_accuracy": 0.19997620582580566,
      "num_tokens": 30668948.0,
      "step": 16665
    },
    {
      "entropy": 5.180909824371338,
      "epoch": 1.392415636485132,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004809341573019929,
      "loss": 4.899,
      "mean_token_accuracy": 0.2146259978413582,
      "num_tokens": 30678818.0,
      "step": 16670
    },
    {
      "entropy": 5.042481327056885,
      "epoch": 1.392833277647845,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004809221624547911,
      "loss": 4.8424,
      "mean_token_accuracy": 0.2095116540789604,
      "num_tokens": 30687731.0,
      "step": 16675
    },
    {
      "entropy": 5.1589202880859375,
      "epoch": 1.393250918810558,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000480910164002674,
      "loss": 4.9477,
      "mean_token_accuracy": 0.20552807599306105,
      "num_tokens": 30696880.0,
      "step": 16680
    },
    {
      "entropy": 5.180277490615845,
      "epoch": 1.393668559973271,
      "grad_norm": 1.4375,
      "learning_rate": 0.00048089816194585153,
      "loss": 4.9324,
      "mean_token_accuracy": 0.20528421849012374,
      "num_tokens": 30706274.0,
      "step": 16685
    },
    {
      "entropy": 5.14558253288269,
      "epoch": 1.394086201135984,
      "grad_norm": 1.375,
      "learning_rate": 0.0004808861562845339,
      "loss": 4.923,
      "mean_token_accuracy": 0.2026342511177063,
      "num_tokens": 30715184.0,
      "step": 16690
    },
    {
      "entropy": 5.161658811569214,
      "epoch": 1.3945038422986968,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004808741470189313,
      "loss": 4.8927,
      "mean_token_accuracy": 0.21110229790210724,
      "num_tokens": 30724501.0,
      "step": 16695
    },
    {
      "entropy": 5.119596815109253,
      "epoch": 1.39492148346141,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004808621341492541,
      "loss": 4.8894,
      "mean_token_accuracy": 0.2052340880036354,
      "num_tokens": 30733496.0,
      "step": 16700
    },
    {
      "entropy": 5.051064300537109,
      "epoch": 1.395339124624123,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004808501176757122,
      "loss": 4.8765,
      "mean_token_accuracy": 0.21282353550195693,
      "num_tokens": 30742096.0,
      "step": 16705
    },
    {
      "entropy": 5.053646087646484,
      "epoch": 1.395756765786836,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004808380975985163,
      "loss": 4.8388,
      "mean_token_accuracy": 0.21058370172977448,
      "num_tokens": 30751452.0,
      "step": 16710
    },
    {
      "entropy": 5.180178308486939,
      "epoch": 1.396174406949549,
      "grad_norm": 3.734375,
      "learning_rate": 0.0004808260739178768,
      "loss": 4.9736,
      "mean_token_accuracy": 0.2056932583451271,
      "num_tokens": 30760686.0,
      "step": 16715
    },
    {
      "entropy": 5.177733945846557,
      "epoch": 1.396592048112262,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004808140466340041,
      "loss": 4.9658,
      "mean_token_accuracy": 0.2056984931230545,
      "num_tokens": 30769374.0,
      "step": 16720
    },
    {
      "entropy": 5.180169105529785,
      "epoch": 1.397009689274975,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004808020157471087,
      "loss": 4.9992,
      "mean_token_accuracy": 0.20220376402139664,
      "num_tokens": 30778910.0,
      "step": 16725
    },
    {
      "entropy": 5.144395637512207,
      "epoch": 1.3974273304376879,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00048078998125740136,
      "loss": 4.9384,
      "mean_token_accuracy": 0.20565304309129714,
      "num_tokens": 30787844.0,
      "step": 16730
    },
    {
      "entropy": 5.1759744644165036,
      "epoch": 1.3978449716004009,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004807779431650927,
      "loss": 4.9987,
      "mean_token_accuracy": 0.20528428852558137,
      "num_tokens": 30797415.0,
      "step": 16735
    },
    {
      "entropy": 5.18585205078125,
      "epoch": 1.398262612763114,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004807659014703934,
      "loss": 4.9612,
      "mean_token_accuracy": 0.2072196677327156,
      "num_tokens": 30805595.0,
      "step": 16740
    },
    {
      "entropy": 5.172084140777588,
      "epoch": 1.398680253925827,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004807538561735143,
      "loss": 4.885,
      "mean_token_accuracy": 0.21559761911630632,
      "num_tokens": 30813840.0,
      "step": 16745
    },
    {
      "entropy": 5.165805912017822,
      "epoch": 1.39909789508854,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00048074180727466636,
      "loss": 4.8863,
      "mean_token_accuracy": 0.20973984748125077,
      "num_tokens": 30822208.0,
      "step": 16750
    },
    {
      "entropy": 5.1997333526611325,
      "epoch": 1.3995155362512528,
      "grad_norm": 1.1875,
      "learning_rate": 0.00048072975477406036,
      "loss": 5.0836,
      "mean_token_accuracy": 0.19651467353105545,
      "num_tokens": 30831804.0,
      "step": 16755
    },
    {
      "entropy": 5.174748420715332,
      "epoch": 1.399933177413966,
      "grad_norm": 1.25,
      "learning_rate": 0.0004807176986719074,
      "loss": 4.9399,
      "mean_token_accuracy": 0.2073385164141655,
      "num_tokens": 30841450.0,
      "step": 16760
    },
    {
      "entropy": 5.168543910980224,
      "epoch": 1.4003508185766789,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00048070563896841843,
      "loss": 4.9814,
      "mean_token_accuracy": 0.20406438410282135,
      "num_tokens": 30849906.0,
      "step": 16765
    },
    {
      "entropy": 5.132986259460449,
      "epoch": 1.400768459739392,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00048069357566380474,
      "loss": 4.9761,
      "mean_token_accuracy": 0.20372409820556642,
      "num_tokens": 30859643.0,
      "step": 16770
    },
    {
      "entropy": 5.107336759567261,
      "epoch": 1.401186100902105,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00048068150875827737,
      "loss": 4.8484,
      "mean_token_accuracy": 0.21198418885469436,
      "num_tokens": 30868728.0,
      "step": 16775
    },
    {
      "entropy": 5.1844603538513185,
      "epoch": 1.401603742064818,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00048066943825204764,
      "loss": 4.9043,
      "mean_token_accuracy": 0.20807045698165894,
      "num_tokens": 30877253.0,
      "step": 16780
    },
    {
      "entropy": 5.147322797775269,
      "epoch": 1.402021383227531,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004806573641453268,
      "loss": 4.9517,
      "mean_token_accuracy": 0.2051568940281868,
      "num_tokens": 30886118.0,
      "step": 16785
    },
    {
      "entropy": 5.114688110351563,
      "epoch": 1.4024390243902438,
      "grad_norm": 1.15625,
      "learning_rate": 0.00048064528643832627,
      "loss": 4.9637,
      "mean_token_accuracy": 0.2068949043750763,
      "num_tokens": 30894533.0,
      "step": 16790
    },
    {
      "entropy": 5.165314102172852,
      "epoch": 1.4028566655529568,
      "grad_norm": 1.21875,
      "learning_rate": 0.00048063320513125747,
      "loss": 4.8225,
      "mean_token_accuracy": 0.2090175837278366,
      "num_tokens": 30904934.0,
      "step": 16795
    },
    {
      "entropy": 5.121867942810058,
      "epoch": 1.4032743067156699,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004806211202243319,
      "loss": 4.8909,
      "mean_token_accuracy": 0.21484961956739426,
      "num_tokens": 30913641.0,
      "step": 16800
    },
    {
      "entropy": 5.127582788467407,
      "epoch": 1.403691947878383,
      "grad_norm": 1.328125,
      "learning_rate": 0.00048060903171776113,
      "loss": 4.8404,
      "mean_token_accuracy": 0.2096607655286789,
      "num_tokens": 30922529.0,
      "step": 16805
    },
    {
      "entropy": 5.118891716003418,
      "epoch": 1.404109589041096,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004805969396117567,
      "loss": 4.9314,
      "mean_token_accuracy": 0.20426862984895705,
      "num_tokens": 30932059.0,
      "step": 16810
    },
    {
      "entropy": 5.206481838226319,
      "epoch": 1.4045272302038088,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00048058484390653037,
      "loss": 5.0632,
      "mean_token_accuracy": 0.199225515127182,
      "num_tokens": 30941141.0,
      "step": 16815
    },
    {
      "entropy": 5.24321346282959,
      "epoch": 1.404944871366522,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004805727446022939,
      "loss": 5.0001,
      "mean_token_accuracy": 0.20443961173295974,
      "num_tokens": 30949785.0,
      "step": 16820
    },
    {
      "entropy": 5.146825456619263,
      "epoch": 1.4053625125292348,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00048056064169925904,
      "loss": 4.8933,
      "mean_token_accuracy": 0.20917191952466965,
      "num_tokens": 30958421.0,
      "step": 16825
    },
    {
      "entropy": 5.025789451599121,
      "epoch": 1.4057801536919479,
      "grad_norm": 1.203125,
      "learning_rate": 0.00048054853519763786,
      "loss": 4.8528,
      "mean_token_accuracy": 0.20913597643375398,
      "num_tokens": 30968177.0,
      "step": 16830
    },
    {
      "entropy": 5.173900699615478,
      "epoch": 1.406197794854661,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004805364250976421,
      "loss": 4.9469,
      "mean_token_accuracy": 0.202530112862587,
      "num_tokens": 30978048.0,
      "step": 16835
    },
    {
      "entropy": 5.126824903488159,
      "epoch": 1.406615436017374,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004805243113994837,
      "loss": 4.9528,
      "mean_token_accuracy": 0.20244967490434645,
      "num_tokens": 30988588.0,
      "step": 16840
    },
    {
      "entropy": 5.2357096672058105,
      "epoch": 1.407033077180087,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004805121941033749,
      "loss": 4.9499,
      "mean_token_accuracy": 0.2031401589512825,
      "num_tokens": 30998000.0,
      "step": 16845
    },
    {
      "entropy": 5.159774398803711,
      "epoch": 1.4074507183427998,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004805000732095278,
      "loss": 4.9671,
      "mean_token_accuracy": 0.2017010197043419,
      "num_tokens": 31007103.0,
      "step": 16850
    },
    {
      "entropy": 5.139955615997314,
      "epoch": 1.4078683595055128,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004804879487181546,
      "loss": 4.9488,
      "mean_token_accuracy": 0.20338874459266662,
      "num_tokens": 31016658.0,
      "step": 16855
    },
    {
      "entropy": 5.242132043838501,
      "epoch": 1.4082860006682258,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004804758206294675,
      "loss": 5.0679,
      "mean_token_accuracy": 0.1973028674721718,
      "num_tokens": 31026968.0,
      "step": 16860
    },
    {
      "entropy": 5.093621730804443,
      "epoch": 1.4087036418309389,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004804636889436787,
      "loss": 4.8692,
      "mean_token_accuracy": 0.2061703845858574,
      "num_tokens": 31035890.0,
      "step": 16865
    },
    {
      "entropy": 5.2286376953125,
      "epoch": 1.409121282993652,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048045155366100085,
      "loss": 5.0351,
      "mean_token_accuracy": 0.20017098188400267,
      "num_tokens": 31045203.0,
      "step": 16870
    },
    {
      "entropy": 5.239011001586914,
      "epoch": 1.4095389241563647,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004804394147816462,
      "loss": 5.0266,
      "mean_token_accuracy": 0.19922225326299667,
      "num_tokens": 31055802.0,
      "step": 16875
    },
    {
      "entropy": 5.19396939277649,
      "epoch": 1.409956565319078,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048042727230582726,
      "loss": 4.9098,
      "mean_token_accuracy": 0.20611011534929274,
      "num_tokens": 31065434.0,
      "step": 16880
    },
    {
      "entropy": 5.098274850845337,
      "epoch": 1.4103742064817908,
      "grad_norm": 1.125,
      "learning_rate": 0.0004804151262337568,
      "loss": 4.8234,
      "mean_token_accuracy": 0.21628083735704423,
      "num_tokens": 31073931.0,
      "step": 16885
    },
    {
      "entropy": 5.179300165176391,
      "epoch": 1.4107918476445038,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004804029765656471,
      "loss": 5.0427,
      "mean_token_accuracy": 0.1911280259490013,
      "num_tokens": 31083068.0,
      "step": 16890
    },
    {
      "entropy": 5.103931283950805,
      "epoch": 1.4112094888072169,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004803908233017111,
      "loss": 4.835,
      "mean_token_accuracy": 0.21857009828090668,
      "num_tokens": 31092306.0,
      "step": 16895
    },
    {
      "entropy": 5.112173128128052,
      "epoch": 1.41162712996993,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004803786664421616,
      "loss": 4.8742,
      "mean_token_accuracy": 0.2151429235935211,
      "num_tokens": 31100540.0,
      "step": 16900
    },
    {
      "entropy": 5.142665195465088,
      "epoch": 1.412044771132643,
      "grad_norm": 1.15625,
      "learning_rate": 0.00048036650598721125,
      "loss": 4.8941,
      "mean_token_accuracy": 0.20315370410680772,
      "num_tokens": 31110337.0,
      "step": 16905
    },
    {
      "entropy": 5.0284308910369875,
      "epoch": 1.4124624122953557,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00048035434193707306,
      "loss": 4.8056,
      "mean_token_accuracy": 0.2059733659029007,
      "num_tokens": 31119985.0,
      "step": 16910
    },
    {
      "entropy": 4.999893140792847,
      "epoch": 1.4128800534580688,
      "grad_norm": 1.296875,
      "learning_rate": 0.00048034217429195986,
      "loss": 4.7692,
      "mean_token_accuracy": 0.2123061388731003,
      "num_tokens": 31128147.0,
      "step": 16915
    },
    {
      "entropy": 5.193738222122192,
      "epoch": 1.4132976946207818,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00048033000305208485,
      "loss": 4.9853,
      "mean_token_accuracy": 0.20356916785240173,
      "num_tokens": 31137150.0,
      "step": 16920
    },
    {
      "entropy": 5.188693904876709,
      "epoch": 1.4137153357834948,
      "grad_norm": 1.328125,
      "learning_rate": 0.00048031782821766086,
      "loss": 4.8822,
      "mean_token_accuracy": 0.20934448689222335,
      "num_tokens": 31145847.0,
      "step": 16925
    },
    {
      "entropy": 5.165197324752808,
      "epoch": 1.4141329769462079,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00048030564978890117,
      "loss": 4.9716,
      "mean_token_accuracy": 0.2056901529431343,
      "num_tokens": 31154313.0,
      "step": 16930
    },
    {
      "entropy": 5.180772542953491,
      "epoch": 1.4145506181089207,
      "grad_norm": 1.7734375,
      "learning_rate": 0.000480293467766019,
      "loss": 5.0431,
      "mean_token_accuracy": 0.199504092335701,
      "num_tokens": 31164104.0,
      "step": 16935
    },
    {
      "entropy": 5.122453308105468,
      "epoch": 1.414968259271634,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004802812821492276,
      "loss": 4.8194,
      "mean_token_accuracy": 0.22000714987516404,
      "num_tokens": 31172833.0,
      "step": 16940
    },
    {
      "entropy": 5.130866479873657,
      "epoch": 1.4153859004343468,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00048026909293874016,
      "loss": 5.0013,
      "mean_token_accuracy": 0.19589665979146959,
      "num_tokens": 31182557.0,
      "step": 16945
    },
    {
      "entropy": 5.172497844696045,
      "epoch": 1.4158035415970598,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004802569001347702,
      "loss": 4.9448,
      "mean_token_accuracy": 0.20659666657447814,
      "num_tokens": 31191641.0,
      "step": 16950
    },
    {
      "entropy": 5.074954223632813,
      "epoch": 1.4162211827597728,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00048024470373753114,
      "loss": 4.8884,
      "mean_token_accuracy": 0.20741498917341233,
      "num_tokens": 31200689.0,
      "step": 16955
    },
    {
      "entropy": 5.222265005111694,
      "epoch": 1.4166388239224859,
      "grad_norm": 1.203125,
      "learning_rate": 0.00048023250374723656,
      "loss": 5.0463,
      "mean_token_accuracy": 0.19864531457424164,
      "num_tokens": 31209745.0,
      "step": 16960
    },
    {
      "entropy": 5.202592849731445,
      "epoch": 1.417056465085199,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004802203001640998,
      "loss": 4.8848,
      "mean_token_accuracy": 0.20685312598943711,
      "num_tokens": 31218802.0,
      "step": 16965
    },
    {
      "entropy": 5.1082252025604244,
      "epoch": 1.4174741062479117,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004802080929883348,
      "loss": 4.8944,
      "mean_token_accuracy": 0.21113411486148834,
      "num_tokens": 31227650.0,
      "step": 16970
    },
    {
      "entropy": 5.092708683013916,
      "epoch": 1.4178917474106247,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004801958822201551,
      "loss": 4.8828,
      "mean_token_accuracy": 0.21509323865175248,
      "num_tokens": 31236490.0,
      "step": 16975
    },
    {
      "entropy": 5.191828918457031,
      "epoch": 1.4183093885733378,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00048018366785977445,
      "loss": 4.9064,
      "mean_token_accuracy": 0.20954085737466813,
      "num_tokens": 31245557.0,
      "step": 16980
    },
    {
      "entropy": 5.082806968688965,
      "epoch": 1.4187270297360508,
      "grad_norm": 1.359375,
      "learning_rate": 0.00048017144990740673,
      "loss": 4.8143,
      "mean_token_accuracy": 0.21702978909015655,
      "num_tokens": 31253804.0,
      "step": 16985
    },
    {
      "entropy": 5.15596570968628,
      "epoch": 1.4191446708987638,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004801592283632658,
      "loss": 4.9348,
      "mean_token_accuracy": 0.2045835018157959,
      "num_tokens": 31262913.0,
      "step": 16990
    },
    {
      "entropy": 5.162899780273437,
      "epoch": 1.4195623120614766,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00048014700322756567,
      "loss": 4.9067,
      "mean_token_accuracy": 0.20777372270822525,
      "num_tokens": 31271274.0,
      "step": 16995
    },
    {
      "entropy": 5.164398241043091,
      "epoch": 1.41997995322419,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004801347745005203,
      "loss": 4.8818,
      "mean_token_accuracy": 0.20761991590261458,
      "num_tokens": 31280599.0,
      "step": 17000
    },
    {
      "entropy": 5.175174951553345,
      "epoch": 1.4203975943869027,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004801225421823437,
      "loss": 4.9236,
      "mean_token_accuracy": 0.20552688390016555,
      "num_tokens": 31288986.0,
      "step": 17005
    },
    {
      "entropy": 5.203184318542481,
      "epoch": 1.4208152355496158,
      "grad_norm": 1.453125,
      "learning_rate": 0.00048011030627325015,
      "loss": 4.9497,
      "mean_token_accuracy": 0.20760907232761383,
      "num_tokens": 31298979.0,
      "step": 17010
    },
    {
      "entropy": 5.106667518615723,
      "epoch": 1.4212328767123288,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00048009806677345394,
      "loss": 4.8814,
      "mean_token_accuracy": 0.20750773549079896,
      "num_tokens": 31307502.0,
      "step": 17015
    },
    {
      "entropy": 5.053784561157227,
      "epoch": 1.4216505178750418,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000480085823683169,
      "loss": 4.8676,
      "mean_token_accuracy": 0.21192067712545395,
      "num_tokens": 31316146.0,
      "step": 17020
    },
    {
      "entropy": 5.178310394287109,
      "epoch": 1.4220681590377549,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004800735770026099,
      "loss": 4.9491,
      "mean_token_accuracy": 0.20472717732191087,
      "num_tokens": 31325542.0,
      "step": 17025
    },
    {
      "entropy": 5.219383621215821,
      "epoch": 1.4224858002004677,
      "grad_norm": 1.203125,
      "learning_rate": 0.000480061326731991,
      "loss": 4.9749,
      "mean_token_accuracy": 0.209326171875,
      "num_tokens": 31334718.0,
      "step": 17030
    },
    {
      "entropy": 5.159142875671387,
      "epoch": 1.4229034413631807,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004800490728715267,
      "loss": 4.9557,
      "mean_token_accuracy": 0.20592936128377914,
      "num_tokens": 31343635.0,
      "step": 17035
    },
    {
      "entropy": 5.141614723205566,
      "epoch": 1.4233210825258937,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004800368154214317,
      "loss": 4.9324,
      "mean_token_accuracy": 0.21271113008260728,
      "num_tokens": 31352488.0,
      "step": 17040
    },
    {
      "entropy": 5.18542447090149,
      "epoch": 1.4237387236886068,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004800245543819203,
      "loss": 4.9917,
      "mean_token_accuracy": 0.20492485761642457,
      "num_tokens": 31361788.0,
      "step": 17045
    },
    {
      "entropy": 5.146648263931274,
      "epoch": 1.4241563648513198,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00048001228975320736,
      "loss": 4.9869,
      "mean_token_accuracy": 0.1987429991364479,
      "num_tokens": 31371190.0,
      "step": 17050
    },
    {
      "entropy": 5.1568560123443605,
      "epoch": 1.4245740060140326,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004800000215355075,
      "loss": 4.9837,
      "mean_token_accuracy": 0.20072211921215058,
      "num_tokens": 31380696.0,
      "step": 17055
    },
    {
      "entropy": 5.267510938644409,
      "epoch": 1.4249916471767459,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004799877497290355,
      "loss": 4.9998,
      "mean_token_accuracy": 0.19768076390028,
      "num_tokens": 31390129.0,
      "step": 17060
    },
    {
      "entropy": 5.171603059768676,
      "epoch": 1.4254092883394587,
      "grad_norm": 1.28125,
      "learning_rate": 0.00047997547433400615,
      "loss": 4.9205,
      "mean_token_accuracy": 0.2044382080435753,
      "num_tokens": 31399697.0,
      "step": 17065
    },
    {
      "entropy": 5.113142967224121,
      "epoch": 1.4258269295021717,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004799631953506345,
      "loss": 4.92,
      "mean_token_accuracy": 0.2030289426445961,
      "num_tokens": 31408873.0,
      "step": 17070
    },
    {
      "entropy": 5.078351974487305,
      "epoch": 1.4262445706648847,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047995091277913535,
      "loss": 4.836,
      "mean_token_accuracy": 0.21095700412988663,
      "num_tokens": 31416980.0,
      "step": 17075
    },
    {
      "entropy": 5.238522863388061,
      "epoch": 1.4266622118275978,
      "grad_norm": 1.3125,
      "learning_rate": 0.00047993862661972373,
      "loss": 5.0141,
      "mean_token_accuracy": 0.19796279221773147,
      "num_tokens": 31425759.0,
      "step": 17080
    },
    {
      "entropy": 5.24111385345459,
      "epoch": 1.4270798529903108,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004799263368726148,
      "loss": 5.0479,
      "mean_token_accuracy": 0.2021453857421875,
      "num_tokens": 31435481.0,
      "step": 17085
    },
    {
      "entropy": 5.183825206756592,
      "epoch": 1.4274974941530236,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004799140435380237,
      "loss": 4.8749,
      "mean_token_accuracy": 0.20228869765996932,
      "num_tokens": 31443924.0,
      "step": 17090
    },
    {
      "entropy": 5.1953691959381105,
      "epoch": 1.4279151353157367,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047990174661616556,
      "loss": 5.0181,
      "mean_token_accuracy": 0.19781327694654466,
      "num_tokens": 31453982.0,
      "step": 17095
    },
    {
      "entropy": 5.179477596282959,
      "epoch": 1.4283327764784497,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047988944610725574,
      "loss": 4.9638,
      "mean_token_accuracy": 0.20740670412778855,
      "num_tokens": 31463292.0,
      "step": 17100
    },
    {
      "entropy": 5.1210143089294435,
      "epoch": 1.4287504176411627,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004798771420115095,
      "loss": 4.9208,
      "mean_token_accuracy": 0.2092076301574707,
      "num_tokens": 31472314.0,
      "step": 17105
    },
    {
      "entropy": 5.204214239120484,
      "epoch": 1.4291680588038758,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004798648343291423,
      "loss": 4.9519,
      "mean_token_accuracy": 0.2046432837843895,
      "num_tokens": 31482231.0,
      "step": 17110
    },
    {
      "entropy": 5.096951866149903,
      "epoch": 1.4295856999665886,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047985252306036966,
      "loss": 4.849,
      "mean_token_accuracy": 0.2106482654809952,
      "num_tokens": 31491405.0,
      "step": 17115
    },
    {
      "entropy": 5.17721152305603,
      "epoch": 1.4300033411293018,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047984020820540695,
      "loss": 5.0044,
      "mean_token_accuracy": 0.1967616617679596,
      "num_tokens": 31501662.0,
      "step": 17120
    },
    {
      "entropy": 5.246503114700317,
      "epoch": 1.4304209822920146,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047982788976446984,
      "loss": 4.9588,
      "mean_token_accuracy": 0.20279151499271392,
      "num_tokens": 31510671.0,
      "step": 17125
    },
    {
      "entropy": 5.166549730300903,
      "epoch": 1.4308386234547277,
      "grad_norm": 1.3125,
      "learning_rate": 0.000479815567737774,
      "loss": 4.8764,
      "mean_token_accuracy": 0.215415920317173,
      "num_tokens": 31519309.0,
      "step": 17130
    },
    {
      "entropy": 5.147456264495849,
      "epoch": 1.4312562646174407,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000479803242125535,
      "loss": 4.9041,
      "mean_token_accuracy": 0.2042377069592476,
      "num_tokens": 31528540.0,
      "step": 17135
    },
    {
      "entropy": 5.173732328414917,
      "epoch": 1.4316739057801537,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00047979091292796886,
      "loss": 4.9668,
      "mean_token_accuracy": 0.20871566534042357,
      "num_tokens": 31538035.0,
      "step": 17140
    },
    {
      "entropy": 5.206404829025269,
      "epoch": 1.4320915469428668,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004797785801452911,
      "loss": 5.0082,
      "mean_token_accuracy": 0.20380700379610062,
      "num_tokens": 31548197.0,
      "step": 17145
    },
    {
      "entropy": 5.199960517883301,
      "epoch": 1.4325091881055796,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004797662437777179,
      "loss": 4.9657,
      "mean_token_accuracy": 0.2136477217078209,
      "num_tokens": 31557268.0,
      "step": 17150
    },
    {
      "entropy": 5.183418130874633,
      "epoch": 1.4329268292682926,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047975390382546523,
      "loss": 4.9817,
      "mean_token_accuracy": 0.20185731649398803,
      "num_tokens": 31566503.0,
      "step": 17155
    },
    {
      "entropy": 5.1704151153564455,
      "epoch": 1.4333444704310057,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047974156028874894,
      "loss": 4.9592,
      "mean_token_accuracy": 0.20368808805942534,
      "num_tokens": 31575754.0,
      "step": 17160
    },
    {
      "entropy": 5.170295190811157,
      "epoch": 1.4337621115937187,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004797292131677851,
      "loss": 4.9901,
      "mean_token_accuracy": 0.19916316866874695,
      "num_tokens": 31585172.0,
      "step": 17165
    },
    {
      "entropy": 5.100339078903199,
      "epoch": 1.4341797527564317,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047971686246279014,
      "loss": 5.0119,
      "mean_token_accuracy": 0.20038705468177795,
      "num_tokens": 31594011.0,
      "step": 17170
    },
    {
      "entropy": 5.135582065582275,
      "epoch": 1.4345973939191445,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004797045081739799,
      "loss": 4.8131,
      "mean_token_accuracy": 0.2142893224954605,
      "num_tokens": 31602598.0,
      "step": 17175
    },
    {
      "entropy": 5.179408836364746,
      "epoch": 1.4350150350818578,
      "grad_norm": 1.25,
      "learning_rate": 0.0004796921503015709,
      "loss": 4.9543,
      "mean_token_accuracy": 0.20100803226232528,
      "num_tokens": 31611589.0,
      "step": 17180
    },
    {
      "entropy": 5.15817608833313,
      "epoch": 1.4354326762445706,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00047967978884577944,
      "loss": 4.8941,
      "mean_token_accuracy": 0.20521107017993928,
      "num_tokens": 31620502.0,
      "step": 17185
    },
    {
      "entropy": 5.2077391147613525,
      "epoch": 1.4358503174072836,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004796674238068219,
      "loss": 5.0172,
      "mean_token_accuracy": 0.19079897701740264,
      "num_tokens": 31630209.0,
      "step": 17190
    },
    {
      "entropy": 5.2305844783782955,
      "epoch": 1.4362679585699967,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047965505518491474,
      "loss": 4.9617,
      "mean_token_accuracy": 0.19922198802232743,
      "num_tokens": 31639346.0,
      "step": 17195
    },
    {
      "entropy": 5.193344497680664,
      "epoch": 1.4366855997327097,
      "grad_norm": 1.390625,
      "learning_rate": 0.00047964268298027456,
      "loss": 4.9658,
      "mean_token_accuracy": 0.19744507372379302,
      "num_tokens": 31647482.0,
      "step": 17200
    },
    {
      "entropy": 5.122843217849732,
      "epoch": 1.4371032408954227,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047963030719311777,
      "loss": 4.9144,
      "mean_token_accuracy": 0.21041752099990846,
      "num_tokens": 31656218.0,
      "step": 17205
    },
    {
      "entropy": 5.242578220367432,
      "epoch": 1.4375208820581356,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004796179278236612,
      "loss": 4.97,
      "mean_token_accuracy": 0.19217330515384673,
      "num_tokens": 31665306.0,
      "step": 17210
    },
    {
      "entropy": 5.115113210678101,
      "epoch": 1.4379385232208486,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004796055448721215,
      "loss": 4.8423,
      "mean_token_accuracy": 0.21090618371963502,
      "num_tokens": 31673689.0,
      "step": 17215
    },
    {
      "entropy": 5.152494287490844,
      "epoch": 1.4383561643835616,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047959315833871546,
      "loss": 5.0411,
      "mean_token_accuracy": 0.19773354530334472,
      "num_tokens": 31683742.0,
      "step": 17220
    },
    {
      "entropy": 5.257281875610351,
      "epoch": 1.4387738055462747,
      "grad_norm": 1.34375,
      "learning_rate": 0.00047958076822365994,
      "loss": 4.9919,
      "mean_token_accuracy": 0.20422667115926743,
      "num_tokens": 31693721.0,
      "step": 17225
    },
    {
      "entropy": 5.1774797439575195,
      "epoch": 1.4391914467089877,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047956837452717185,
      "loss": 4.9053,
      "mean_token_accuracy": 0.21812136918306352,
      "num_tokens": 31703103.0,
      "step": 17230
    },
    {
      "entropy": 5.0958949565887455,
      "epoch": 1.4396090878717005,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047955597724946803,
      "loss": 4.83,
      "mean_token_accuracy": 0.20905085206031798,
      "num_tokens": 31711718.0,
      "step": 17235
    },
    {
      "entropy": 5.138310956954956,
      "epoch": 1.4400267290344138,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004795435763907656,
      "loss": 4.9715,
      "mean_token_accuracy": 0.2112927556037903,
      "num_tokens": 31721248.0,
      "step": 17240
    },
    {
      "entropy": 5.128442001342774,
      "epoch": 1.4404443701971266,
      "grad_norm": 1.25,
      "learning_rate": 0.0004795311719512817,
      "loss": 4.9511,
      "mean_token_accuracy": 0.20305074900388717,
      "num_tokens": 31730109.0,
      "step": 17245
    },
    {
      "entropy": 5.3257392883300785,
      "epoch": 1.4408620113598396,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047951876393123344,
      "loss": 5.1864,
      "mean_token_accuracy": 0.18572192192077636,
      "num_tokens": 31739768.0,
      "step": 17250
    },
    {
      "entropy": 5.272120666503906,
      "epoch": 1.4412796525225526,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00047950635233083805,
      "loss": 5.086,
      "mean_token_accuracy": 0.19837736934423447,
      "num_tokens": 31749965.0,
      "step": 17255
    },
    {
      "entropy": 5.206469535827637,
      "epoch": 1.4416972936852657,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047949393715031275,
      "loss": 4.8657,
      "mean_token_accuracy": 0.21432308554649354,
      "num_tokens": 31759112.0,
      "step": 17260
    },
    {
      "entropy": 5.160031318664551,
      "epoch": 1.4421149348479787,
      "grad_norm": 1.265625,
      "learning_rate": 0.00047948151838987505,
      "loss": 4.891,
      "mean_token_accuracy": 0.2101273164153099,
      "num_tokens": 31768535.0,
      "step": 17265
    },
    {
      "entropy": 5.084855842590332,
      "epoch": 1.4425325760106915,
      "grad_norm": 1.328125,
      "learning_rate": 0.00047946909604974205,
      "loss": 4.8786,
      "mean_token_accuracy": 0.21449561715126036,
      "num_tokens": 31777088.0,
      "step": 17270
    },
    {
      "entropy": 5.080096292495727,
      "epoch": 1.4429502171734045,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004794566701301315,
      "loss": 4.8302,
      "mean_token_accuracy": 0.21047762930393218,
      "num_tokens": 31785895.0,
      "step": 17275
    },
    {
      "entropy": 5.162518548965454,
      "epoch": 1.4433678583361176,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00047944424063126083,
      "loss": 4.9983,
      "mean_token_accuracy": 0.20339642614126205,
      "num_tokens": 31795404.0,
      "step": 17280
    },
    {
      "entropy": 5.204475736618042,
      "epoch": 1.4437854994988306,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004794318075533477,
      "loss": 5.0132,
      "mean_token_accuracy": 0.19441054314374923,
      "num_tokens": 31804899.0,
      "step": 17285
    },
    {
      "entropy": 5.124831438064575,
      "epoch": 1.4442031406615436,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004794193708966096,
      "loss": 4.897,
      "mean_token_accuracy": 0.20802751630544664,
      "num_tokens": 31814055.0,
      "step": 17290
    },
    {
      "entropy": 5.07937183380127,
      "epoch": 1.4446207818242565,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047940693066126445,
      "loss": 4.9057,
      "mean_token_accuracy": 0.216205957531929,
      "num_tokens": 31823193.0,
      "step": 17295
    },
    {
      "entropy": 5.173711824417114,
      "epoch": 1.4450384229869697,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004793944868475299,
      "loss": 4.9282,
      "mean_token_accuracy": 0.20527011752128602,
      "num_tokens": 31832359.0,
      "step": 17300
    },
    {
      "entropy": 5.167520666122437,
      "epoch": 1.4454560641496825,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004793820394556238,
      "loss": 4.8965,
      "mean_token_accuracy": 0.2064964011311531,
      "num_tokens": 31841477.0,
      "step": 17305
    },
    {
      "entropy": 5.208763551712036,
      "epoch": 1.4458737053123956,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004793695884857641,
      "loss": 4.9948,
      "mean_token_accuracy": 0.1982489287853241,
      "num_tokens": 31851068.0,
      "step": 17310
    },
    {
      "entropy": 5.152752876281738,
      "epoch": 1.4462913464751086,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004793571339381687,
      "loss": 4.8703,
      "mean_token_accuracy": 0.20902798622846602,
      "num_tokens": 31860026.0,
      "step": 17315
    },
    {
      "entropy": 5.20547399520874,
      "epoch": 1.4467089876378216,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004793446758130558,
      "loss": 4.9165,
      "mean_token_accuracy": 0.20233047902584075,
      "num_tokens": 31868654.0,
      "step": 17320
    },
    {
      "entropy": 5.109862279891968,
      "epoch": 1.4471266288005347,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047933221411064334,
      "loss": 4.8955,
      "mean_token_accuracy": 0.2100677013397217,
      "num_tokens": 31877801.0,
      "step": 17325
    },
    {
      "entropy": 5.121067380905151,
      "epoch": 1.4475442699632475,
      "grad_norm": 1.125,
      "learning_rate": 0.0004793197488311494,
      "loss": 4.9444,
      "mean_token_accuracy": 0.20043278634548187,
      "num_tokens": 31887266.0,
      "step": 17330
    },
    {
      "entropy": 5.152824640274048,
      "epoch": 1.4479619111259605,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047930727997479253,
      "loss": 4.8498,
      "mean_token_accuracy": 0.21673823446035384,
      "num_tokens": 31895957.0,
      "step": 17335
    },
    {
      "entropy": 5.195457458496094,
      "epoch": 1.4483795522886735,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004792948075417906,
      "loss": 5.025,
      "mean_token_accuracy": 0.20296918004751205,
      "num_tokens": 31904443.0,
      "step": 17340
    },
    {
      "entropy": 5.108335828781128,
      "epoch": 1.4487971934513866,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004792823315323622,
      "loss": 4.9007,
      "mean_token_accuracy": 0.2038506582379341,
      "num_tokens": 31913534.0,
      "step": 17345
    },
    {
      "entropy": 5.221584320068359,
      "epoch": 1.4492148346140996,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004792698519467258,
      "loss": 4.9965,
      "mean_token_accuracy": 0.20276074409484862,
      "num_tokens": 31922905.0,
      "step": 17350
    },
    {
      "entropy": 5.222964525222778,
      "epoch": 1.4496324757768126,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004792573687850997,
      "loss": 4.9945,
      "mean_token_accuracy": 0.19795973896980285,
      "num_tokens": 31932861.0,
      "step": 17355
    },
    {
      "entropy": 5.203044271469116,
      "epoch": 1.4500501169395257,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004792448820477025,
      "loss": 4.9646,
      "mean_token_accuracy": 0.1985362097620964,
      "num_tokens": 31941100.0,
      "step": 17360
    },
    {
      "entropy": 5.128841972351074,
      "epoch": 1.4504677581022385,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004792323917347528,
      "loss": 4.9384,
      "mean_token_accuracy": 0.208122119307518,
      "num_tokens": 31949860.0,
      "step": 17365
    },
    {
      "entropy": 5.087614059448242,
      "epoch": 1.4508853992649515,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004792198978464692,
      "loss": 4.8293,
      "mean_token_accuracy": 0.20738860666751863,
      "num_tokens": 31958216.0,
      "step": 17370
    },
    {
      "entropy": 5.108155393600464,
      "epoch": 1.4513030404276646,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004792074003830705,
      "loss": 4.8892,
      "mean_token_accuracy": 0.21115265637636185,
      "num_tokens": 31966967.0,
      "step": 17375
    },
    {
      "entropy": 5.141740894317627,
      "epoch": 1.4517206815903776,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004791948993447754,
      "loss": 4.9203,
      "mean_token_accuracy": 0.20578179955482484,
      "num_tokens": 31976852.0,
      "step": 17380
    },
    {
      "entropy": 5.157869720458985,
      "epoch": 1.4521383227530906,
      "grad_norm": 1.375,
      "learning_rate": 0.0004791823947318029,
      "loss": 4.8624,
      "mean_token_accuracy": 0.2120947778224945,
      "num_tokens": 31985966.0,
      "step": 17385
    },
    {
      "entropy": 5.116045045852661,
      "epoch": 1.4525559639158034,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004791698865443717,
      "loss": 4.8807,
      "mean_token_accuracy": 0.21410396695137024,
      "num_tokens": 31994897.0,
      "step": 17390
    },
    {
      "entropy": 5.138995409011841,
      "epoch": 1.4529736050785165,
      "grad_norm": 1.3125,
      "learning_rate": 0.00047915737478270093,
      "loss": 4.9102,
      "mean_token_accuracy": 0.2024269863963127,
      "num_tokens": 32004504.0,
      "step": 17395
    },
    {
      "entropy": 5.194048547744751,
      "epoch": 1.4533912462412295,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004791448594470095,
      "loss": 4.8912,
      "mean_token_accuracy": 0.2078936979174614,
      "num_tokens": 32012907.0,
      "step": 17400
    },
    {
      "entropy": 5.153306674957276,
      "epoch": 1.4538088874039425,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004791323405375167,
      "loss": 4.8356,
      "mean_token_accuracy": 0.21033525466918945,
      "num_tokens": 32021778.0,
      "step": 17405
    },
    {
      "entropy": 5.150767707824707,
      "epoch": 1.4542265285666556,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004791198180544415,
      "loss": 4.9172,
      "mean_token_accuracy": 0.2056371971964836,
      "num_tokens": 32030471.0,
      "step": 17410
    },
    {
      "entropy": 5.163404226303101,
      "epoch": 1.4546441697293686,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004791072919980032,
      "loss": 5.0353,
      "mean_token_accuracy": 0.1994469240307808,
      "num_tokens": 32039747.0,
      "step": 17415
    },
    {
      "entropy": 5.247016859054566,
      "epoch": 1.4550618108920816,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047909476236842105,
      "loss": 4.9585,
      "mean_token_accuracy": 0.20563313663005828,
      "num_tokens": 32048911.0,
      "step": 17420
    },
    {
      "entropy": 5.217907953262329,
      "epoch": 1.4554794520547945,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004790822291659144,
      "loss": 4.9839,
      "mean_token_accuracy": 0.19313722103834152,
      "num_tokens": 32058942.0,
      "step": 17425
    },
    {
      "entropy": 5.152946424484253,
      "epoch": 1.4558970932175075,
      "grad_norm": 1.859375,
      "learning_rate": 0.00047906969239070276,
      "loss": 5.0238,
      "mean_token_accuracy": 0.20146015733480455,
      "num_tokens": 32067828.0,
      "step": 17430
    },
    {
      "entropy": 5.137798309326172,
      "epoch": 1.4563147343802205,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047905715204300543,
      "loss": 4.8815,
      "mean_token_accuracy": 0.21326787918806075,
      "num_tokens": 32077765.0,
      "step": 17435
    },
    {
      "entropy": 5.1397802352905275,
      "epoch": 1.4567323755429336,
      "grad_norm": 1.359375,
      "learning_rate": 0.000479044608123042,
      "loss": 4.9161,
      "mean_token_accuracy": 0.21037512868642808,
      "num_tokens": 32087525.0,
      "step": 17440
    },
    {
      "entropy": 5.243380355834961,
      "epoch": 1.4571500167056466,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004790320606310322,
      "loss": 5.0485,
      "mean_token_accuracy": 0.1998554527759552,
      "num_tokens": 32097005.0,
      "step": 17445
    },
    {
      "entropy": 5.261539363861084,
      "epoch": 1.4575676578683594,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004790195095671956,
      "loss": 5.0067,
      "mean_token_accuracy": 0.19786621183156966,
      "num_tokens": 32106761.0,
      "step": 17450
    },
    {
      "entropy": 5.1976401805877686,
      "epoch": 1.4579852990310724,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004790069549317518,
      "loss": 4.9143,
      "mean_token_accuracy": 0.20135050117969513,
      "num_tokens": 32115803.0,
      "step": 17455
    },
    {
      "entropy": 4.979160118103027,
      "epoch": 1.4584029401937855,
      "grad_norm": 1.25,
      "learning_rate": 0.0004789943967249207,
      "loss": 4.8408,
      "mean_token_accuracy": 0.2161891609430313,
      "num_tokens": 32124889.0,
      "step": 17460
    },
    {
      "entropy": 5.064678001403808,
      "epoch": 1.4588205813564985,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004789818349469222,
      "loss": 4.9203,
      "mean_token_accuracy": 0.21207238137722015,
      "num_tokens": 32133765.0,
      "step": 17465
    },
    {
      "entropy": 5.211964893341064,
      "epoch": 1.4592382225192115,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047896926959797615,
      "loss": 4.9651,
      "mean_token_accuracy": 0.20454254001379013,
      "num_tokens": 32142890.0,
      "step": 17470
    },
    {
      "entropy": 5.1365180015563965,
      "epoch": 1.4596558636819246,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004789567006783025,
      "loss": 4.8983,
      "mean_token_accuracy": 0.2055329591035843,
      "num_tokens": 32151791.0,
      "step": 17475
    },
    {
      "entropy": 5.113968992233277,
      "epoch": 1.4600735048446376,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004789441281881213,
      "loss": 4.9398,
      "mean_token_accuracy": 0.20932377129793167,
      "num_tokens": 32160313.0,
      "step": 17480
    },
    {
      "entropy": 5.193583679199219,
      "epoch": 1.4604911460073504,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004789315521276526,
      "loss": 4.908,
      "mean_token_accuracy": 0.20491093695163726,
      "num_tokens": 32170044.0,
      "step": 17485
    },
    {
      "entropy": 5.2609593868255615,
      "epoch": 1.4609087871700634,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004789189724971167,
      "loss": 4.9861,
      "mean_token_accuracy": 0.20303948372602462,
      "num_tokens": 32178907.0,
      "step": 17490
    },
    {
      "entropy": 5.164228677749634,
      "epoch": 1.4613264283327765,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047890638929673366,
      "loss": 4.9074,
      "mean_token_accuracy": 0.2001961126923561,
      "num_tokens": 32188197.0,
      "step": 17495
    },
    {
      "entropy": 5.167341327667236,
      "epoch": 1.4617440694954895,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004788938025267238,
      "loss": 4.9896,
      "mean_token_accuracy": 0.20144791752099991,
      "num_tokens": 32198206.0,
      "step": 17500
    },
    {
      "entropy": 5.30652084350586,
      "epoch": 1.4621617106582026,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047888121218730756,
      "loss": 5.0706,
      "mean_token_accuracy": 0.20111843198537827,
      "num_tokens": 32207306.0,
      "step": 17505
    },
    {
      "entropy": 5.1538004875183105,
      "epoch": 1.4625793518209154,
      "grad_norm": 1.265625,
      "learning_rate": 0.00047886861827870533,
      "loss": 4.9089,
      "mean_token_accuracy": 0.20871967524290086,
      "num_tokens": 32215945.0,
      "step": 17510
    },
    {
      "entropy": 5.071648788452149,
      "epoch": 1.4629969929836284,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004788560208011375,
      "loss": 4.8394,
      "mean_token_accuracy": 0.20934180468320845,
      "num_tokens": 32224601.0,
      "step": 17515
    },
    {
      "entropy": 5.077131986618042,
      "epoch": 1.4634146341463414,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004788434197548246,
      "loss": 4.8134,
      "mean_token_accuracy": 0.21695306450128554,
      "num_tokens": 32233265.0,
      "step": 17520
    },
    {
      "entropy": 5.081355142593384,
      "epoch": 1.4638322753090545,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004788308151399873,
      "loss": 4.8308,
      "mean_token_accuracy": 0.21558058261871338,
      "num_tokens": 32242749.0,
      "step": 17525
    },
    {
      "entropy": 5.166075134277344,
      "epoch": 1.4642499164717675,
      "grad_norm": 1.28125,
      "learning_rate": 0.00047881820695684624,
      "loss": 4.9361,
      "mean_token_accuracy": 0.2084842935204506,
      "num_tokens": 32251295.0,
      "step": 17530
    },
    {
      "entropy": 5.088173580169678,
      "epoch": 1.4646675576344805,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004788055952056221,
      "loss": 4.9171,
      "mean_token_accuracy": 0.2037196159362793,
      "num_tokens": 32261099.0,
      "step": 17535
    },
    {
      "entropy": 5.143040323257447,
      "epoch": 1.4650851987971936,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00047879297988653576,
      "loss": 4.9709,
      "mean_token_accuracy": 0.20258133709430695,
      "num_tokens": 32269420.0,
      "step": 17540
    },
    {
      "entropy": 5.166236734390258,
      "epoch": 1.4655028399599064,
      "grad_norm": 1.3125,
      "learning_rate": 0.00047878036099980795,
      "loss": 4.8611,
      "mean_token_accuracy": 0.210283300280571,
      "num_tokens": 32278455.0,
      "step": 17545
    },
    {
      "entropy": 5.187725162506103,
      "epoch": 1.4659204811226194,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047876773854565964,
      "loss": 4.9872,
      "mean_token_accuracy": 0.20197107642889023,
      "num_tokens": 32287292.0,
      "step": 17550
    },
    {
      "entropy": 5.2117067813873295,
      "epoch": 1.4663381222853324,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004787551125243118,
      "loss": 5.0174,
      "mean_token_accuracy": 0.19835789501667023,
      "num_tokens": 32296384.0,
      "step": 17555
    },
    {
      "entropy": 5.065980768203735,
      "epoch": 1.4667557634480455,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004787424829359854,
      "loss": 4.9084,
      "mean_token_accuracy": 0.21653362065553666,
      "num_tokens": 32304908.0,
      "step": 17560
    },
    {
      "entropy": 5.19198751449585,
      "epoch": 1.4671734046107585,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004787298497809017,
      "loss": 4.9253,
      "mean_token_accuracy": 0.21038949191570283,
      "num_tokens": 32314843.0,
      "step": 17565
    },
    {
      "entropy": 5.160279941558838,
      "epoch": 1.4675910457734713,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004787172130592817,
      "loss": 4.8329,
      "mean_token_accuracy": 0.21287367939949037,
      "num_tokens": 32324101.0,
      "step": 17570
    },
    {
      "entropy": 5.10477204322815,
      "epoch": 1.4680086869361844,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00047870457277134667,
      "loss": 4.9215,
      "mean_token_accuracy": 0.21111877411603927,
      "num_tokens": 32333736.0,
      "step": 17575
    },
    {
      "entropy": 5.156060743331909,
      "epoch": 1.4684263280988974,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004786919289173179,
      "loss": 4.9526,
      "mean_token_accuracy": 0.20638263076543809,
      "num_tokens": 32342369.0,
      "step": 17580
    },
    {
      "entropy": 5.156100416183472,
      "epoch": 1.4688439692616104,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047867928149741666,
      "loss": 4.9275,
      "mean_token_accuracy": 0.2125802993774414,
      "num_tokens": 32351576.0,
      "step": 17585
    },
    {
      "entropy": 5.2604821681976315,
      "epoch": 1.4692616104243235,
      "grad_norm": 1.453125,
      "learning_rate": 0.00047866663051186444,
      "loss": 4.9613,
      "mean_token_accuracy": 0.2024817317724228,
      "num_tokens": 32360355.0,
      "step": 17590
    },
    {
      "entropy": 5.176391220092773,
      "epoch": 1.4696792515870365,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004786539759608828,
      "loss": 4.9616,
      "mean_token_accuracy": 0.20698162615299226,
      "num_tokens": 32368611.0,
      "step": 17595
    },
    {
      "entropy": 5.201526021957397,
      "epoch": 1.4700968927497495,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047864131784469314,
      "loss": 4.9887,
      "mean_token_accuracy": 0.20488526076078414,
      "num_tokens": 32377753.0,
      "step": 17600
    },
    {
      "entropy": 5.226001596450805,
      "epoch": 1.4705145339124623,
      "grad_norm": 1.203125,
      "learning_rate": 0.000478628656163517,
      "loss": 5.0589,
      "mean_token_accuracy": 0.19451992809772492,
      "num_tokens": 32388135.0,
      "step": 17605
    },
    {
      "entropy": 5.161518716812134,
      "epoch": 1.4709321750751754,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004786159909175762,
      "loss": 4.9409,
      "mean_token_accuracy": 0.21178358942270278,
      "num_tokens": 32397149.0,
      "step": 17610
    },
    {
      "entropy": 5.089328908920288,
      "epoch": 1.4713498162378884,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004786033221070924,
      "loss": 4.9008,
      "mean_token_accuracy": 0.20821199864149093,
      "num_tokens": 32405811.0,
      "step": 17615
    },
    {
      "entropy": 5.201108455657959,
      "epoch": 1.4717674574006014,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004785906497322873,
      "loss": 4.8711,
      "mean_token_accuracy": 0.21389972418546677,
      "num_tokens": 32413800.0,
      "step": 17620
    },
    {
      "entropy": 5.13468017578125,
      "epoch": 1.4721850985633145,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047857797379338284,
      "loss": 4.9271,
      "mean_token_accuracy": 0.2137158066034317,
      "num_tokens": 32422462.0,
      "step": 17625
    },
    {
      "entropy": 5.151666641235352,
      "epoch": 1.4726027397260273,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047856529429060097,
      "loss": 5.0109,
      "mean_token_accuracy": 0.1993408367037773,
      "num_tokens": 32433149.0,
      "step": 17630
    },
    {
      "entropy": 5.1210814952850345,
      "epoch": 1.4730203808887403,
      "grad_norm": 1.25,
      "learning_rate": 0.00047855261122416355,
      "loss": 4.8886,
      "mean_token_accuracy": 0.2108069971203804,
      "num_tokens": 32442033.0,
      "step": 17635
    },
    {
      "entropy": 5.236160707473755,
      "epoch": 1.4734380220514534,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004785399245942926,
      "loss": 4.9621,
      "mean_token_accuracy": 0.20034821033477784,
      "num_tokens": 32451189.0,
      "step": 17640
    },
    {
      "entropy": 5.164788007736206,
      "epoch": 1.4738556632141664,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004785272344012104,
      "loss": 4.9049,
      "mean_token_accuracy": 0.2090345785021782,
      "num_tokens": 32459929.0,
      "step": 17645
    },
    {
      "entropy": 5.052565050125122,
      "epoch": 1.4742733043768794,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004785145406451389,
      "loss": 4.9657,
      "mean_token_accuracy": 0.20289636850357057,
      "num_tokens": 32468275.0,
      "step": 17650
    },
    {
      "entropy": 5.138612413406372,
      "epoch": 1.4746909455395925,
      "grad_norm": 1.25,
      "learning_rate": 0.0004785018433263004,
      "loss": 4.9293,
      "mean_token_accuracy": 0.21013051569461821,
      "num_tokens": 32477833.0,
      "step": 17655
    },
    {
      "entropy": 5.136714029312134,
      "epoch": 1.4751085867023055,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047848914244491715,
      "loss": 4.8372,
      "mean_token_accuracy": 0.21268783807754515,
      "num_tokens": 32486458.0,
      "step": 17660
    },
    {
      "entropy": 5.131064605712891,
      "epoch": 1.4755262278650183,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004784764380012115,
      "loss": 4.888,
      "mean_token_accuracy": 0.21557111144065857,
      "num_tokens": 32495355.0,
      "step": 17665
    },
    {
      "entropy": 5.122048854827881,
      "epoch": 1.4759438690277313,
      "grad_norm": 1.3125,
      "learning_rate": 0.00047846372999540596,
      "loss": 4.8665,
      "mean_token_accuracy": 0.20972199738025665,
      "num_tokens": 32504083.0,
      "step": 17670
    },
    {
      "entropy": 5.11871371269226,
      "epoch": 1.4763615101904444,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047845101842772285,
      "loss": 4.9951,
      "mean_token_accuracy": 0.20441833585500718,
      "num_tokens": 32513303.0,
      "step": 17675
    },
    {
      "entropy": 5.15890212059021,
      "epoch": 1.4767791513531574,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047843830329838475,
      "loss": 4.9631,
      "mean_token_accuracy": 0.203650626540184,
      "num_tokens": 32522183.0,
      "step": 17680
    },
    {
      "entropy": 5.24750542640686,
      "epoch": 1.4771967925158704,
      "grad_norm": 1.125,
      "learning_rate": 0.00047842558460761437,
      "loss": 4.9896,
      "mean_token_accuracy": 0.19980325996875764,
      "num_tokens": 32531443.0,
      "step": 17685
    },
    {
      "entropy": 5.206852006912231,
      "epoch": 1.4776144336785833,
      "grad_norm": 1.265625,
      "learning_rate": 0.00047841286235563417,
      "loss": 4.8834,
      "mean_token_accuracy": 0.21466579288244247,
      "num_tokens": 32539435.0,
      "step": 17690
    },
    {
      "entropy": 5.030299234390259,
      "epoch": 1.4780320748412963,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00047840013654266707,
      "loss": 4.9135,
      "mean_token_accuracy": 0.20987793803215027,
      "num_tokens": 32548465.0,
      "step": 17695
    },
    {
      "entropy": 5.0974523544311525,
      "epoch": 1.4784497160040093,
      "grad_norm": 1.25,
      "learning_rate": 0.0004783874071689357,
      "loss": 4.8965,
      "mean_token_accuracy": 0.21612953692674636,
      "num_tokens": 32556511.0,
      "step": 17700
    },
    {
      "entropy": 5.23331937789917,
      "epoch": 1.4788673571667224,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047837467423466284,
      "loss": 5.0646,
      "mean_token_accuracy": 0.2012180656194687,
      "num_tokens": 32566077.0,
      "step": 17705
    },
    {
      "entropy": 5.260114955902099,
      "epoch": 1.4792849983294354,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047836193774007164,
      "loss": 5.0118,
      "mean_token_accuracy": 0.2024947941303253,
      "num_tokens": 32575942.0,
      "step": 17710
    },
    {
      "entropy": 5.113724946975708,
      "epoch": 1.4797026394921484,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004783491976853849,
      "loss": 4.8918,
      "mean_token_accuracy": 0.21939975172281265,
      "num_tokens": 32584595.0,
      "step": 17715
    },
    {
      "entropy": 5.152707767486572,
      "epoch": 1.4801202806548615,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047833645407082564,
      "loss": 4.8824,
      "mean_token_accuracy": 0.2027769938111305,
      "num_tokens": 32594048.0,
      "step": 17720
    },
    {
      "entropy": 5.089960241317749,
      "epoch": 1.4805379218175743,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004783237068966171,
      "loss": 4.7722,
      "mean_token_accuracy": 0.20870449841022493,
      "num_tokens": 32603197.0,
      "step": 17725
    },
    {
      "entropy": 5.084911251068116,
      "epoch": 1.4809555629802873,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047831095616298225,
      "loss": 4.9071,
      "mean_token_accuracy": 0.20294980853796005,
      "num_tokens": 32612882.0,
      "step": 17730
    },
    {
      "entropy": 5.18469877243042,
      "epoch": 1.4813732041430003,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004782982018701444,
      "loss": 4.9205,
      "mean_token_accuracy": 0.2096533104777336,
      "num_tokens": 32622425.0,
      "step": 17735
    },
    {
      "entropy": 5.301807260513305,
      "epoch": 1.4817908453057134,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004782854440183268,
      "loss": 5.003,
      "mean_token_accuracy": 0.2053149625658989,
      "num_tokens": 32631295.0,
      "step": 17740
    },
    {
      "entropy": 5.267035913467407,
      "epoch": 1.4822084864684264,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004782726826077528,
      "loss": 5.0087,
      "mean_token_accuracy": 0.1982637509703636,
      "num_tokens": 32639476.0,
      "step": 17745
    },
    {
      "entropy": 5.145189952850342,
      "epoch": 1.4826261276311392,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047825991763864587,
      "loss": 4.9271,
      "mean_token_accuracy": 0.20669638961553574,
      "num_tokens": 32648704.0,
      "step": 17750
    },
    {
      "entropy": 5.188858270645142,
      "epoch": 1.4830437687938522,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047824714911122934,
      "loss": 5.0215,
      "mean_token_accuracy": 0.201498281955719,
      "num_tokens": 32658258.0,
      "step": 17755
    },
    {
      "entropy": 5.19865345954895,
      "epoch": 1.4834614099565653,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00047823437702572684,
      "loss": 4.9611,
      "mean_token_accuracy": 0.20220645815134047,
      "num_tokens": 32668002.0,
      "step": 17760
    },
    {
      "entropy": 5.190167856216431,
      "epoch": 1.4838790511192783,
      "grad_norm": 1.265625,
      "learning_rate": 0.000478221601382362,
      "loss": 5.0185,
      "mean_token_accuracy": 0.1991311565041542,
      "num_tokens": 32677961.0,
      "step": 17765
    },
    {
      "entropy": 5.200455856323242,
      "epoch": 1.4842966922819913,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047820882218135835,
      "loss": 4.8822,
      "mean_token_accuracy": 0.2139560729265213,
      "num_tokens": 32687523.0,
      "step": 17770
    },
    {
      "entropy": 5.110609436035157,
      "epoch": 1.4847143334447044,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047819603942293964,
      "loss": 4.8739,
      "mean_token_accuracy": 0.2057161122560501,
      "num_tokens": 32696563.0,
      "step": 17775
    },
    {
      "entropy": 5.202744102478027,
      "epoch": 1.4851319746074174,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004781832531073297,
      "loss": 4.9462,
      "mean_token_accuracy": 0.2016652047634125,
      "num_tokens": 32704928.0,
      "step": 17780
    },
    {
      "entropy": 5.205867719650269,
      "epoch": 1.4855496157701302,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00047817046323475246,
      "loss": 4.9712,
      "mean_token_accuracy": 0.20048919022083284,
      "num_tokens": 32714228.0,
      "step": 17785
    },
    {
      "entropy": 5.1866391658782955,
      "epoch": 1.4859672569328433,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004781576698054316,
      "loss": 4.9307,
      "mean_token_accuracy": 0.20085130929946898,
      "num_tokens": 32724164.0,
      "step": 17790
    },
    {
      "entropy": 5.119447898864746,
      "epoch": 1.4863848980955563,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004781448728195912,
      "loss": 4.9248,
      "mean_token_accuracy": 0.21305187046527863,
      "num_tokens": 32732516.0,
      "step": 17795
    },
    {
      "entropy": 5.197665786743164,
      "epoch": 1.4868025392582693,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047813207227745527,
      "loss": 5.087,
      "mean_token_accuracy": 0.1914411574602127,
      "num_tokens": 32741819.0,
      "step": 17800
    },
    {
      "entropy": 5.214570236206055,
      "epoch": 1.4872201804209824,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047811926817924785,
      "loss": 4.8668,
      "mean_token_accuracy": 0.22073378711938857,
      "num_tokens": 32751255.0,
      "step": 17805
    },
    {
      "entropy": 5.304936075210572,
      "epoch": 1.4876378215836952,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047810646052519324,
      "loss": 5.0127,
      "mean_token_accuracy": 0.20148788541555404,
      "num_tokens": 32760275.0,
      "step": 17810
    },
    {
      "entropy": 5.179897308349609,
      "epoch": 1.4880554627464082,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004780936493155155,
      "loss": 4.9325,
      "mean_token_accuracy": 0.2110305294394493,
      "num_tokens": 32769146.0,
      "step": 17815
    },
    {
      "entropy": 5.14980845451355,
      "epoch": 1.4884731039091212,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00047808083455043893,
      "loss": 5.0382,
      "mean_token_accuracy": 0.20028142184019088,
      "num_tokens": 32778523.0,
      "step": 17820
    },
    {
      "entropy": 5.155358791351318,
      "epoch": 1.4888907450718343,
      "grad_norm": 1.25,
      "learning_rate": 0.00047806801623018804,
      "loss": 4.8008,
      "mean_token_accuracy": 0.21446058452129363,
      "num_tokens": 32786746.0,
      "step": 17825
    },
    {
      "entropy": 5.153324127197266,
      "epoch": 1.4893083862345473,
      "grad_norm": 1.34375,
      "learning_rate": 0.000478055194354987,
      "loss": 4.9022,
      "mean_token_accuracy": 0.21454654335975648,
      "num_tokens": 32796396.0,
      "step": 17830
    },
    {
      "entropy": 5.0984845638275145,
      "epoch": 1.4897260273972603,
      "grad_norm": 1.265625,
      "learning_rate": 0.00047804236892506034,
      "loss": 4.9544,
      "mean_token_accuracy": 0.20779917687177657,
      "num_tokens": 32806329.0,
      "step": 17835
    },
    {
      "entropy": 5.132034587860107,
      "epoch": 1.4901436685599734,
      "grad_norm": 1.328125,
      "learning_rate": 0.00047802953994063266,
      "loss": 4.9106,
      "mean_token_accuracy": 0.20635858327150344,
      "num_tokens": 32815959.0,
      "step": 17840
    },
    {
      "entropy": 5.212565994262695,
      "epoch": 1.4905613097226862,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047801670740192843,
      "loss": 4.879,
      "mean_token_accuracy": 0.20789871364831924,
      "num_tokens": 32824762.0,
      "step": 17845
    },
    {
      "entropy": 5.060662317276001,
      "epoch": 1.4909789508853992,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047800387130917245,
      "loss": 4.8304,
      "mean_token_accuracy": 0.2123759150505066,
      "num_tokens": 32833949.0,
      "step": 17850
    },
    {
      "entropy": 5.175077486038208,
      "epoch": 1.4913965920481123,
      "grad_norm": 1.46875,
      "learning_rate": 0.00047799103166258926,
      "loss": 4.9606,
      "mean_token_accuracy": 0.20414691120386125,
      "num_tokens": 32843428.0,
      "step": 17855
    },
    {
      "entropy": 5.340870380401611,
      "epoch": 1.4918142332108253,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047797818846240383,
      "loss": 5.0964,
      "mean_token_accuracy": 0.19391078054904937,
      "num_tokens": 32853878.0,
      "step": 17860
    },
    {
      "entropy": 5.179860591888428,
      "epoch": 1.4922318743735383,
      "grad_norm": 1.375,
      "learning_rate": 0.0004779653417088408,
      "loss": 4.9288,
      "mean_token_accuracy": 0.1996622920036316,
      "num_tokens": 32863011.0,
      "step": 17865
    },
    {
      "entropy": 5.159111785888672,
      "epoch": 1.4926495155362511,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047795249140212523,
      "loss": 4.9449,
      "mean_token_accuracy": 0.20661493241786957,
      "num_tokens": 32872238.0,
      "step": 17870
    },
    {
      "entropy": 5.1725325107574465,
      "epoch": 1.4930671566989642,
      "grad_norm": 1.390625,
      "learning_rate": 0.00047793963754248195,
      "loss": 4.9291,
      "mean_token_accuracy": 0.20572706311941147,
      "num_tokens": 32880382.0,
      "step": 17875
    },
    {
      "entropy": 5.223012399673462,
      "epoch": 1.4934847978616772,
      "grad_norm": 1.28125,
      "learning_rate": 0.000477926780130136,
      "loss": 4.9463,
      "mean_token_accuracy": 0.20631791800260543,
      "num_tokens": 32889199.0,
      "step": 17880
    },
    {
      "entropy": 5.259348726272583,
      "epoch": 1.4939024390243902,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047791391916531255,
      "loss": 5.0806,
      "mean_token_accuracy": 0.1934283196926117,
      "num_tokens": 32898867.0,
      "step": 17885
    },
    {
      "entropy": 5.27960696220398,
      "epoch": 1.4943200801871033,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004779010546482367,
      "loss": 5.0762,
      "mean_token_accuracy": 0.19310788363218306,
      "num_tokens": 32907947.0,
      "step": 17890
    },
    {
      "entropy": 5.224570226669312,
      "epoch": 1.4947377213498163,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004778881865791337,
      "loss": 4.8941,
      "mean_token_accuracy": 0.20317959636449814,
      "num_tokens": 32917045.0,
      "step": 17895
    },
    {
      "entropy": 5.176211738586426,
      "epoch": 1.4951553625125293,
      "grad_norm": 4.40625,
      "learning_rate": 0.00047787531495822866,
      "loss": 4.8696,
      "mean_token_accuracy": 0.20931265354156495,
      "num_tokens": 32926644.0,
      "step": 17900
    },
    {
      "entropy": 5.160429430007935,
      "epoch": 1.4955730036752422,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047786243978574706,
      "loss": 4.9535,
      "mean_token_accuracy": 0.2008269414305687,
      "num_tokens": 32935920.0,
      "step": 17905
    },
    {
      "entropy": 5.207805204391479,
      "epoch": 1.4959906448379552,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047784956106191435,
      "loss": 4.9464,
      "mean_token_accuracy": 0.2034326434135437,
      "num_tokens": 32945561.0,
      "step": 17910
    },
    {
      "entropy": 5.180936527252197,
      "epoch": 1.4964082860006682,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004778366787869559,
      "loss": 4.8924,
      "mean_token_accuracy": 0.20784412771463395,
      "num_tokens": 32955531.0,
      "step": 17915
    },
    {
      "entropy": 5.084345483779908,
      "epoch": 1.4968259271633813,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004778237929610971,
      "loss": 4.7361,
      "mean_token_accuracy": 0.21543285250663757,
      "num_tokens": 32964001.0,
      "step": 17920
    },
    {
      "entropy": 5.208854818344117,
      "epoch": 1.4972435683260943,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004778109035845638,
      "loss": 4.9982,
      "mean_token_accuracy": 0.20459900945425033,
      "num_tokens": 32973707.0,
      "step": 17925
    },
    {
      "entropy": 5.232402276992798,
      "epoch": 1.497661209488807,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047779801065758146,
      "loss": 4.9612,
      "mean_token_accuracy": 0.20442092716693877,
      "num_tokens": 32983029.0,
      "step": 17930
    },
    {
      "entropy": 5.21531400680542,
      "epoch": 1.4980788506515201,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004777851141803758,
      "loss": 4.9398,
      "mean_token_accuracy": 0.20492665767669677,
      "num_tokens": 32992351.0,
      "step": 17935
    },
    {
      "entropy": 5.091637325286865,
      "epoch": 1.4984964918142332,
      "grad_norm": 1.25,
      "learning_rate": 0.0004777722141531727,
      "loss": 4.7866,
      "mean_token_accuracy": 0.22111931443214417,
      "num_tokens": 33001090.0,
      "step": 17940
    },
    {
      "entropy": 5.146827220916748,
      "epoch": 1.4989141329769462,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047775931057619786,
      "loss": 5.0131,
      "mean_token_accuracy": 0.1949007347226143,
      "num_tokens": 33010294.0,
      "step": 17945
    },
    {
      "entropy": 5.184524917602539,
      "epoch": 1.4993317741396592,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004777464034496773,
      "loss": 4.9376,
      "mean_token_accuracy": 0.20777415931224824,
      "num_tokens": 33018633.0,
      "step": 17950
    },
    {
      "entropy": 5.155618333816529,
      "epoch": 1.4997494153023723,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004777334927738368,
      "loss": 4.9887,
      "mean_token_accuracy": 0.20271705687046052,
      "num_tokens": 33028291.0,
      "step": 17955
    },
    {
      "entropy": 5.1880943775177,
      "epoch": 1.5001670564650853,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047772057854890247,
      "loss": 4.8972,
      "mean_token_accuracy": 0.20541446208953856,
      "num_tokens": 33036603.0,
      "step": 17960
    },
    {
      "entropy": 5.206019687652588,
      "epoch": 1.5005846976277981,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047770766077510045,
      "loss": 4.9338,
      "mean_token_accuracy": 0.2093082293868065,
      "num_tokens": 33045329.0,
      "step": 17965
    },
    {
      "entropy": 5.118571949005127,
      "epoch": 1.5010023387905111,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047769473945265685,
      "loss": 4.8814,
      "mean_token_accuracy": 0.20696995556354522,
      "num_tokens": 33054753.0,
      "step": 17970
    },
    {
      "entropy": 5.109956932067871,
      "epoch": 1.5014199799532242,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004776818145817979,
      "loss": 4.8962,
      "mean_token_accuracy": 0.21865332573652269,
      "num_tokens": 33063469.0,
      "step": 17975
    },
    {
      "entropy": 5.180388355255127,
      "epoch": 1.5018376211159372,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047766888616274973,
      "loss": 4.9633,
      "mean_token_accuracy": 0.20262960642576217,
      "num_tokens": 33072247.0,
      "step": 17980
    },
    {
      "entropy": 5.205013799667358,
      "epoch": 1.5022552622786502,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047765595419573876,
      "loss": 4.9792,
      "mean_token_accuracy": 0.2068116009235382,
      "num_tokens": 33083505.0,
      "step": 17985
    },
    {
      "entropy": 5.155536079406739,
      "epoch": 1.502672903441363,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047764301868099136,
      "loss": 4.9749,
      "mean_token_accuracy": 0.20611559152603148,
      "num_tokens": 33094104.0,
      "step": 17990
    },
    {
      "entropy": 5.121548748016357,
      "epoch": 1.5030905446040763,
      "grad_norm": 1.390625,
      "learning_rate": 0.00047763007961873407,
      "loss": 4.9089,
      "mean_token_accuracy": 0.21213555485010147,
      "num_tokens": 33102936.0,
      "step": 17995
    },
    {
      "entropy": 5.158829879760742,
      "epoch": 1.5035081857667891,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004776171370091933,
      "loss": 4.8961,
      "mean_token_accuracy": 0.2133016124367714,
      "num_tokens": 33111648.0,
      "step": 18000
    },
    {
      "epoch": 1.5035081857667891,
      "eval_entropy": 4.979449562742609,
      "eval_loss": 4.952423572540283,
      "eval_mean_token_accuracy": 0.21629016192579037,
      "eval_num_tokens": 33111648.0,
      "eval_runtime": 27.4963,
      "eval_samples_per_second": 1373.385,
      "eval_steps_per_second": 171.696,
      "step": 18000
    },
    {
      "entropy": 5.176357269287109,
      "epoch": 1.5039258269295022,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004776041908525956,
      "loss": 4.9527,
      "mean_token_accuracy": 0.20086924880743026,
      "num_tokens": 33120235.0,
      "step": 18005
    },
    {
      "entropy": 5.123017168045044,
      "epoch": 1.5043434680922152,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004775912411491677,
      "loss": 4.9036,
      "mean_token_accuracy": 0.2126241534948349,
      "num_tokens": 33129152.0,
      "step": 18010
    },
    {
      "entropy": 5.095234060287476,
      "epoch": 1.504761109254928,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004775782878991363,
      "loss": 4.8422,
      "mean_token_accuracy": 0.212987545132637,
      "num_tokens": 33137907.0,
      "step": 18015
    },
    {
      "entropy": 5.124683809280396,
      "epoch": 1.5051787504176413,
      "grad_norm": 1.34375,
      "learning_rate": 0.00047756533110272813,
      "loss": 4.904,
      "mean_token_accuracy": 0.2081712692975998,
      "num_tokens": 33146001.0,
      "step": 18020
    },
    {
      "entropy": 5.181609058380127,
      "epoch": 1.505596391580354,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047755237076016997,
      "loss": 4.9055,
      "mean_token_accuracy": 0.20306651890277863,
      "num_tokens": 33155389.0,
      "step": 18025
    },
    {
      "entropy": 5.1843517303466795,
      "epoch": 1.506014032743067,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047753940687168876,
      "loss": 4.9682,
      "mean_token_accuracy": 0.20933347940444946,
      "num_tokens": 33164965.0,
      "step": 18030
    },
    {
      "entropy": 5.179042387008667,
      "epoch": 1.5064316739057801,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004775264394375114,
      "loss": 5.0059,
      "mean_token_accuracy": 0.20203229635953904,
      "num_tokens": 33174590.0,
      "step": 18035
    },
    {
      "entropy": 5.151254892349243,
      "epoch": 1.5068493150684932,
      "grad_norm": 1.25,
      "learning_rate": 0.000477513468457865,
      "loss": 4.904,
      "mean_token_accuracy": 0.20161326378583908,
      "num_tokens": 33183063.0,
      "step": 18040
    },
    {
      "entropy": 5.214575529098511,
      "epoch": 1.5072669562312062,
      "grad_norm": 1.796875,
      "learning_rate": 0.00047750049393297656,
      "loss": 5.0379,
      "mean_token_accuracy": 0.19787142425775528,
      "num_tokens": 33192198.0,
      "step": 18045
    },
    {
      "entropy": 5.293501281738282,
      "epoch": 1.507684597393919,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047748751586307314,
      "loss": 5.0302,
      "mean_token_accuracy": 0.20143998712301253,
      "num_tokens": 33202445.0,
      "step": 18050
    },
    {
      "entropy": 5.218570709228516,
      "epoch": 1.5081022385566323,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047747453424838216,
      "loss": 4.9447,
      "mean_token_accuracy": 0.21308365762233733,
      "num_tokens": 33211017.0,
      "step": 18055
    },
    {
      "entropy": 5.075514793395996,
      "epoch": 1.508519879719345,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00047746154908913065,
      "loss": 4.9122,
      "mean_token_accuracy": 0.21436023861169815,
      "num_tokens": 33220345.0,
      "step": 18060
    },
    {
      "entropy": 5.144996976852417,
      "epoch": 1.5089375208820581,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000477448560385546,
      "loss": 4.929,
      "mean_token_accuracy": 0.2127312660217285,
      "num_tokens": 33229849.0,
      "step": 18065
    },
    {
      "entropy": 5.159467935562134,
      "epoch": 1.5093551620447712,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004774355681378557,
      "loss": 4.884,
      "mean_token_accuracy": 0.20854503959417342,
      "num_tokens": 33239079.0,
      "step": 18070
    },
    {
      "entropy": 5.136460685729981,
      "epoch": 1.509772803207484,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047742257234628703,
      "loss": 4.8567,
      "mean_token_accuracy": 0.20987775623798371,
      "num_tokens": 33248048.0,
      "step": 18075
    },
    {
      "entropy": 5.095633935928345,
      "epoch": 1.5101904443701972,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047740957301106756,
      "loss": 4.8823,
      "mean_token_accuracy": 0.2191465601325035,
      "num_tokens": 33257173.0,
      "step": 18080
    },
    {
      "entropy": 5.135183334350586,
      "epoch": 1.51060808553291,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000477396570132425,
      "loss": 4.9602,
      "mean_token_accuracy": 0.20581815987825394,
      "num_tokens": 33266800.0,
      "step": 18085
    },
    {
      "entropy": 5.17824125289917,
      "epoch": 1.511025726695623,
      "grad_norm": 1.265625,
      "learning_rate": 0.00047738356371058673,
      "loss": 5.0063,
      "mean_token_accuracy": 0.2004191741347313,
      "num_tokens": 33276061.0,
      "step": 18090
    },
    {
      "entropy": 5.139435052871704,
      "epoch": 1.511443367858336,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004773705537457806,
      "loss": 4.8193,
      "mean_token_accuracy": 0.21633868664503098,
      "num_tokens": 33284786.0,
      "step": 18095
    },
    {
      "entropy": 5.136729764938354,
      "epoch": 1.5118610090210491,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00047735754023823433,
      "loss": 4.9003,
      "mean_token_accuracy": 0.20846717357635497,
      "num_tokens": 33293849.0,
      "step": 18100
    },
    {
      "entropy": 5.06649079322815,
      "epoch": 1.5122786501837622,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004773445231881757,
      "loss": 4.8033,
      "mean_token_accuracy": 0.21076516509056092,
      "num_tokens": 33303387.0,
      "step": 18105
    },
    {
      "entropy": 5.100030994415283,
      "epoch": 1.512696291346475,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047733150259583274,
      "loss": 4.8459,
      "mean_token_accuracy": 0.20906783640384674,
      "num_tokens": 33312502.0,
      "step": 18110
    },
    {
      "entropy": 5.062272310256958,
      "epoch": 1.5131139325091882,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047731847846143325,
      "loss": 4.8595,
      "mean_token_accuracy": 0.2147008717060089,
      "num_tokens": 33321909.0,
      "step": 18115
    },
    {
      "entropy": 5.0757347583770756,
      "epoch": 1.513531573671901,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004773054507852052,
      "loss": 4.8199,
      "mean_token_accuracy": 0.20744241625070572,
      "num_tokens": 33331249.0,
      "step": 18120
    },
    {
      "entropy": 5.122629022598266,
      "epoch": 1.513949214834614,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004772924195673768,
      "loss": 4.9054,
      "mean_token_accuracy": 0.213039730489254,
      "num_tokens": 33340850.0,
      "step": 18125
    },
    {
      "entropy": 5.164441156387329,
      "epoch": 1.5143668559973271,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047727938480817604,
      "loss": 4.9601,
      "mean_token_accuracy": 0.1997520387172699,
      "num_tokens": 33350429.0,
      "step": 18130
    },
    {
      "entropy": 5.188778924942016,
      "epoch": 1.51478449716004,
      "grad_norm": 1.25,
      "learning_rate": 0.00047726634650783117,
      "loss": 4.9355,
      "mean_token_accuracy": 0.2045842781662941,
      "num_tokens": 33359687.0,
      "step": 18135
    },
    {
      "entropy": 5.183815431594849,
      "epoch": 1.5152021383227532,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004772533046665705,
      "loss": 4.9847,
      "mean_token_accuracy": 0.20023991018533707,
      "num_tokens": 33368177.0,
      "step": 18140
    },
    {
      "entropy": 5.118700790405273,
      "epoch": 1.515619779485466,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004772402592846222,
      "loss": 4.9444,
      "mean_token_accuracy": 0.2120738446712494,
      "num_tokens": 33376604.0,
      "step": 18145
    },
    {
      "entropy": 5.116697311401367,
      "epoch": 1.516037420648179,
      "grad_norm": 1.265625,
      "learning_rate": 0.00047722721036221467,
      "loss": 4.8705,
      "mean_token_accuracy": 0.20925575792789458,
      "num_tokens": 33385488.0,
      "step": 18150
    },
    {
      "entropy": 5.18545241355896,
      "epoch": 1.516455061810892,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004772141578995765,
      "loss": 4.9773,
      "mean_token_accuracy": 0.19877718687057494,
      "num_tokens": 33394524.0,
      "step": 18155
    },
    {
      "entropy": 5.177974796295166,
      "epoch": 1.516872702973605,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047720110189693605,
      "loss": 4.9397,
      "mean_token_accuracy": 0.2102349281311035,
      "num_tokens": 33404300.0,
      "step": 18160
    },
    {
      "entropy": 5.131511163711548,
      "epoch": 1.5172903441363181,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047718804235452193,
      "loss": 4.883,
      "mean_token_accuracy": 0.2104376047849655,
      "num_tokens": 33414050.0,
      "step": 18165
    },
    {
      "entropy": 5.129007196426391,
      "epoch": 1.517707985299031,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004771749792725628,
      "loss": 5.0064,
      "mean_token_accuracy": 0.2067019134759903,
      "num_tokens": 33423981.0,
      "step": 18170
    },
    {
      "entropy": 5.12883996963501,
      "epoch": 1.5181256264617442,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004771619126512873,
      "loss": 4.7635,
      "mean_token_accuracy": 0.21658313423395156,
      "num_tokens": 33433104.0,
      "step": 18175
    },
    {
      "entropy": 5.189019346237183,
      "epoch": 1.518543267624457,
      "grad_norm": 1.25,
      "learning_rate": 0.00047714884249092413,
      "loss": 4.9803,
      "mean_token_accuracy": 0.20387874841690062,
      "num_tokens": 33442034.0,
      "step": 18180
    },
    {
      "entropy": 5.10506010055542,
      "epoch": 1.51896090878717,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004771357687917021,
      "loss": 4.8852,
      "mean_token_accuracy": 0.20706933587789536,
      "num_tokens": 33450431.0,
      "step": 18185
    },
    {
      "entropy": 5.160455322265625,
      "epoch": 1.519378549949883,
      "grad_norm": 1.390625,
      "learning_rate": 0.00047712269155385017,
      "loss": 5.0249,
      "mean_token_accuracy": 0.20287092179059982,
      "num_tokens": 33459790.0,
      "step": 18190
    },
    {
      "entropy": 5.232061672210693,
      "epoch": 1.519796191112596,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047710961077759736,
      "loss": 5.0127,
      "mean_token_accuracy": 0.1987200453877449,
      "num_tokens": 33468802.0,
      "step": 18195
    },
    {
      "entropy": 5.17845344543457,
      "epoch": 1.5202138322753092,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047709652646317236,
      "loss": 4.8635,
      "mean_token_accuracy": 0.21232950389385224,
      "num_tokens": 33477756.0,
      "step": 18200
    },
    {
      "entropy": 5.158196067810058,
      "epoch": 1.520631473438022,
      "grad_norm": 1.3125,
      "learning_rate": 0.00047708343861080447,
      "loss": 4.9849,
      "mean_token_accuracy": 0.20510114282369613,
      "num_tokens": 33487410.0,
      "step": 18205
    },
    {
      "entropy": 5.195968437194824,
      "epoch": 1.521049114600735,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004770703472207227,
      "loss": 4.9146,
      "mean_token_accuracy": 0.20780282318592072,
      "num_tokens": 33496682.0,
      "step": 18210
    },
    {
      "entropy": 5.261792182922363,
      "epoch": 1.521466755763448,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004770572522931563,
      "loss": 5.0624,
      "mean_token_accuracy": 0.19593590199947358,
      "num_tokens": 33505754.0,
      "step": 18215
    },
    {
      "entropy": 5.152641439437867,
      "epoch": 1.521884396926161,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004770441538283345,
      "loss": 4.9399,
      "mean_token_accuracy": 0.20675729364156722,
      "num_tokens": 33514904.0,
      "step": 18220
    },
    {
      "entropy": 5.164978694915772,
      "epoch": 1.522302038088874,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047703105182648656,
      "loss": 4.9046,
      "mean_token_accuracy": 0.20947396010160446,
      "num_tokens": 33523684.0,
      "step": 18225
    },
    {
      "entropy": 5.108409214019775,
      "epoch": 1.522719679251587,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004770179462878419,
      "loss": 4.9542,
      "mean_token_accuracy": 0.20460040718317032,
      "num_tokens": 33533066.0,
      "step": 18230
    },
    {
      "entropy": 5.134424209594727,
      "epoch": 1.5231373204143002,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004770048372126299,
      "loss": 4.9091,
      "mean_token_accuracy": 0.19907529056072235,
      "num_tokens": 33542048.0,
      "step": 18235
    },
    {
      "entropy": 5.131352758407592,
      "epoch": 1.523554961577013,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004769917246010801,
      "loss": 4.8337,
      "mean_token_accuracy": 0.21348237991333008,
      "num_tokens": 33550785.0,
      "step": 18240
    },
    {
      "entropy": 5.123459196090698,
      "epoch": 1.523972602739726,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004769786084534219,
      "loss": 4.9245,
      "mean_token_accuracy": 0.19860857874155044,
      "num_tokens": 33560141.0,
      "step": 18245
    },
    {
      "entropy": 5.1856261730194095,
      "epoch": 1.524390243902439,
      "grad_norm": 1.125,
      "learning_rate": 0.00047696548876988517,
      "loss": 4.9585,
      "mean_token_accuracy": 0.20823672860860826,
      "num_tokens": 33568866.0,
      "step": 18250
    },
    {
      "entropy": 5.162366056442261,
      "epoch": 1.5248078850651519,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004769523655506994,
      "loss": 4.9358,
      "mean_token_accuracy": 0.20596401989459992,
      "num_tokens": 33579217.0,
      "step": 18255
    },
    {
      "entropy": 5.2000127792358395,
      "epoch": 1.5252255262278651,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004769392387960944,
      "loss": 5.0222,
      "mean_token_accuracy": 0.1990941658616066,
      "num_tokens": 33589608.0,
      "step": 18260
    },
    {
      "entropy": 5.21013617515564,
      "epoch": 1.525643167390578,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004769261085063,
      "loss": 4.946,
      "mean_token_accuracy": 0.20351733565330504,
      "num_tokens": 33599263.0,
      "step": 18265
    },
    {
      "entropy": 5.061718416213989,
      "epoch": 1.526060808553291,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047691297468154594,
      "loss": 4.9279,
      "mean_token_accuracy": 0.21195425391197203,
      "num_tokens": 33607946.0,
      "step": 18270
    },
    {
      "entropy": 5.189308166503906,
      "epoch": 1.526478449716004,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004768998373220622,
      "loss": 5.0187,
      "mean_token_accuracy": 0.20352350175380707,
      "num_tokens": 33617194.0,
      "step": 18275
    },
    {
      "entropy": 5.206690979003906,
      "epoch": 1.526896090878717,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004768866964280789,
      "loss": 4.9558,
      "mean_token_accuracy": 0.20961373746395112,
      "num_tokens": 33626796.0,
      "step": 18280
    },
    {
      "entropy": 5.168476676940918,
      "epoch": 1.52731373204143,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00047687355199982584,
      "loss": 4.9098,
      "mean_token_accuracy": 0.20307690650224686,
      "num_tokens": 33636290.0,
      "step": 18285
    },
    {
      "entropy": 5.135450792312622,
      "epoch": 1.5277313732041429,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004768604040375333,
      "loss": 4.9309,
      "mean_token_accuracy": 0.21176013052463533,
      "num_tokens": 33644489.0,
      "step": 18290
    },
    {
      "entropy": 5.097818660736084,
      "epoch": 1.5281490143668561,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047684725254143137,
      "loss": 4.818,
      "mean_token_accuracy": 0.21431533992290497,
      "num_tokens": 33653677.0,
      "step": 18295
    },
    {
      "entropy": 5.076485824584961,
      "epoch": 1.528566655529569,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004768340975117504,
      "loss": 4.8868,
      "mean_token_accuracy": 0.2098665252327919,
      "num_tokens": 33662787.0,
      "step": 18300
    },
    {
      "entropy": 5.146691465377808,
      "epoch": 1.528984296692282,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004768209389487205,
      "loss": 4.9383,
      "mean_token_accuracy": 0.22334171533584596,
      "num_tokens": 33671167.0,
      "step": 18305
    },
    {
      "entropy": 5.15317063331604,
      "epoch": 1.529401937854995,
      "grad_norm": 1.125,
      "learning_rate": 0.0004768077768525722,
      "loss": 4.8165,
      "mean_token_accuracy": 0.21480318158864975,
      "num_tokens": 33679253.0,
      "step": 18310
    },
    {
      "entropy": 5.153139877319336,
      "epoch": 1.5298195790177078,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004767946112235359,
      "loss": 4.9338,
      "mean_token_accuracy": 0.20799509882926942,
      "num_tokens": 33688863.0,
      "step": 18315
    },
    {
      "entropy": 5.133846998214722,
      "epoch": 1.530237220180421,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000476781442061842,
      "loss": 4.9189,
      "mean_token_accuracy": 0.20811158716678618,
      "num_tokens": 33697201.0,
      "step": 18320
    },
    {
      "entropy": 5.24550232887268,
      "epoch": 1.5306548613431339,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047676826936772097,
      "loss": 4.9922,
      "mean_token_accuracy": 0.2088351294398308,
      "num_tokens": 33706075.0,
      "step": 18325
    },
    {
      "entropy": 5.172389125823974,
      "epoch": 1.531072502505847,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047675509314140365,
      "loss": 5.0027,
      "mean_token_accuracy": 0.2036559373140335,
      "num_tokens": 33715290.0,
      "step": 18330
    },
    {
      "entropy": 5.11182599067688,
      "epoch": 1.53149014366856,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004767419133831206,
      "loss": 4.8565,
      "mean_token_accuracy": 0.21695488840341567,
      "num_tokens": 33724508.0,
      "step": 18335
    },
    {
      "entropy": 5.216186237335205,
      "epoch": 1.531907784831273,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00047672873009310234,
      "loss": 4.9822,
      "mean_token_accuracy": 0.20474612414836885,
      "num_tokens": 33733333.0,
      "step": 18340
    },
    {
      "entropy": 5.073234033584595,
      "epoch": 1.532325425993986,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00047671554327157996,
      "loss": 4.8522,
      "mean_token_accuracy": 0.21291346698999405,
      "num_tokens": 33741780.0,
      "step": 18345
    },
    {
      "entropy": 5.068338775634766,
      "epoch": 1.5327430671566988,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047670235291878417,
      "loss": 4.828,
      "mean_token_accuracy": 0.21620676219463347,
      "num_tokens": 33750259.0,
      "step": 18350
    },
    {
      "entropy": 5.179843139648438,
      "epoch": 1.533160708319412,
      "grad_norm": 1.25,
      "learning_rate": 0.0004766891590349459,
      "loss": 4.9639,
      "mean_token_accuracy": 0.1983931392431259,
      "num_tokens": 33760383.0,
      "step": 18355
    },
    {
      "entropy": 5.139260959625244,
      "epoch": 1.533578349482125,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004766759616202961,
      "loss": 4.8689,
      "mean_token_accuracy": 0.2125263050198555,
      "num_tokens": 33768524.0,
      "step": 18360
    },
    {
      "entropy": 5.077771329879761,
      "epoch": 1.533995990644838,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004766627606750659,
      "loss": 4.9026,
      "mean_token_accuracy": 0.20900066792964936,
      "num_tokens": 33777448.0,
      "step": 18365
    },
    {
      "entropy": 5.185121250152588,
      "epoch": 1.534413631807551,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004766495561994863,
      "loss": 4.9568,
      "mean_token_accuracy": 0.2094233065843582,
      "num_tokens": 33787316.0,
      "step": 18370
    },
    {
      "entropy": 5.18723316192627,
      "epoch": 1.534831272970264,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004766363481937885,
      "loss": 4.871,
      "mean_token_accuracy": 0.20830435305833817,
      "num_tokens": 33796945.0,
      "step": 18375
    },
    {
      "entropy": 5.204163932800293,
      "epoch": 1.535248914132977,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047662313665820365,
      "loss": 5.0046,
      "mean_token_accuracy": 0.19414145797491072,
      "num_tokens": 33807866.0,
      "step": 18380
    },
    {
      "entropy": 5.126193284988403,
      "epoch": 1.5356665552956899,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004766099215929632,
      "loss": 4.9101,
      "mean_token_accuracy": 0.20640022456645965,
      "num_tokens": 33817298.0,
      "step": 18385
    },
    {
      "entropy": 5.062049531936646,
      "epoch": 1.536084196458403,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047659670299829833,
      "loss": 4.8849,
      "mean_token_accuracy": 0.2144513264298439,
      "num_tokens": 33826552.0,
      "step": 18390
    },
    {
      "entropy": 5.112410306930542,
      "epoch": 1.536501837621116,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004765834808744405,
      "loss": 4.9116,
      "mean_token_accuracy": 0.20876617580652237,
      "num_tokens": 33836800.0,
      "step": 18395
    },
    {
      "entropy": 5.159519910812378,
      "epoch": 1.536919478783829,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047657025522162123,
      "loss": 4.8968,
      "mean_token_accuracy": 0.2128308266401291,
      "num_tokens": 33846245.0,
      "step": 18400
    },
    {
      "entropy": 5.162251710891724,
      "epoch": 1.537337119946542,
      "grad_norm": 1.21875,
      "learning_rate": 0.000476557026040072,
      "loss": 4.8528,
      "mean_token_accuracy": 0.21294583529233932,
      "num_tokens": 33855100.0,
      "step": 18405
    },
    {
      "entropy": 5.071875858306885,
      "epoch": 1.5377547611092548,
      "grad_norm": 1.125,
      "learning_rate": 0.00047654379333002446,
      "loss": 4.7883,
      "mean_token_accuracy": 0.21053078174591064,
      "num_tokens": 33863307.0,
      "step": 18410
    },
    {
      "entropy": 5.126738786697388,
      "epoch": 1.538172402271968,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047653055709171013,
      "loss": 4.9352,
      "mean_token_accuracy": 0.20654311776161194,
      "num_tokens": 33871891.0,
      "step": 18415
    },
    {
      "entropy": 5.092452430725098,
      "epoch": 1.5385900434346809,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047651731732536084,
      "loss": 4.9231,
      "mean_token_accuracy": 0.2092199832201004,
      "num_tokens": 33880913.0,
      "step": 18420
    },
    {
      "entropy": 5.235208320617676,
      "epoch": 1.539007684597394,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047650407403120845,
      "loss": 4.9746,
      "mean_token_accuracy": 0.2042548805475235,
      "num_tokens": 33889838.0,
      "step": 18425
    },
    {
      "entropy": 5.2177670955657955,
      "epoch": 1.539425325760107,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047649082720948455,
      "loss": 4.9705,
      "mean_token_accuracy": 0.20518514961004258,
      "num_tokens": 33898989.0,
      "step": 18430
    },
    {
      "entropy": 5.176686000823975,
      "epoch": 1.53984296692282,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00047647757686042125,
      "loss": 4.9275,
      "mean_token_accuracy": 0.20313141942024232,
      "num_tokens": 33907884.0,
      "step": 18435
    },
    {
      "entropy": 5.177460241317749,
      "epoch": 1.540260608085533,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047646432298425047,
      "loss": 4.9453,
      "mean_token_accuracy": 0.20252098590135575,
      "num_tokens": 33916762.0,
      "step": 18440
    },
    {
      "entropy": 5.133030033111572,
      "epoch": 1.5406782492482458,
      "grad_norm": 1.265625,
      "learning_rate": 0.00047645106558120417,
      "loss": 4.823,
      "mean_token_accuracy": 0.2116239845752716,
      "num_tokens": 33925413.0,
      "step": 18445
    },
    {
      "entropy": 5.082526159286499,
      "epoch": 1.541095890410959,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004764378046515145,
      "loss": 4.8938,
      "mean_token_accuracy": 0.20893584191799164,
      "num_tokens": 33934247.0,
      "step": 18450
    },
    {
      "entropy": 5.09034423828125,
      "epoch": 1.5415135315736719,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004764245401954136,
      "loss": 4.8385,
      "mean_token_accuracy": 0.21551306396722794,
      "num_tokens": 33943610.0,
      "step": 18455
    },
    {
      "entropy": 5.299184417724609,
      "epoch": 1.541931172736385,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004764112722131337,
      "loss": 5.1488,
      "mean_token_accuracy": 0.18970570862293243,
      "num_tokens": 33953447.0,
      "step": 18460
    },
    {
      "entropy": 5.251912355422974,
      "epoch": 1.542348813899098,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047639800070490695,
      "loss": 4.9597,
      "mean_token_accuracy": 0.2061938911676407,
      "num_tokens": 33962702.0,
      "step": 18465
    },
    {
      "entropy": 5.252039766311645,
      "epoch": 1.5427664550618108,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004763847256709658,
      "loss": 4.9687,
      "mean_token_accuracy": 0.20013826489448547,
      "num_tokens": 33972604.0,
      "step": 18470
    },
    {
      "entropy": 5.1424909114837645,
      "epoch": 1.543184096224524,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004763714471115427,
      "loss": 5.0284,
      "mean_token_accuracy": 0.20131392627954484,
      "num_tokens": 33982443.0,
      "step": 18475
    },
    {
      "entropy": 5.152569770812988,
      "epoch": 1.5436017373872368,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004763581650268699,
      "loss": 4.9292,
      "mean_token_accuracy": 0.210069677233696,
      "num_tokens": 33990688.0,
      "step": 18480
    },
    {
      "entropy": 5.224278259277344,
      "epoch": 1.5440193785499499,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047634487941718016,
      "loss": 4.9704,
      "mean_token_accuracy": 0.20423737317323684,
      "num_tokens": 33999657.0,
      "step": 18485
    },
    {
      "entropy": 5.208137130737304,
      "epoch": 1.544437019712663,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004763315902827059,
      "loss": 5.0191,
      "mean_token_accuracy": 0.2027379184961319,
      "num_tokens": 34009842.0,
      "step": 18490
    },
    {
      "entropy": 5.310316658020019,
      "epoch": 1.544854660875376,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047631829762367976,
      "loss": 5.0442,
      "mean_token_accuracy": 0.19166422337293626,
      "num_tokens": 34019600.0,
      "step": 18495
    },
    {
      "entropy": 5.234209203720093,
      "epoch": 1.545272302038089,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004763050014403346,
      "loss": 5.0134,
      "mean_token_accuracy": 0.19782192409038543,
      "num_tokens": 34028256.0,
      "step": 18500
    },
    {
      "entropy": 5.201214838027954,
      "epoch": 1.5456899432008018,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047629170173290296,
      "loss": 4.9516,
      "mean_token_accuracy": 0.2088353231549263,
      "num_tokens": 34037117.0,
      "step": 18505
    },
    {
      "entropy": 5.104575681686401,
      "epoch": 1.546107584363515,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047627839850161784,
      "loss": 4.9241,
      "mean_token_accuracy": 0.2078513666987419,
      "num_tokens": 34045697.0,
      "step": 18510
    },
    {
      "entropy": 5.237617683410645,
      "epoch": 1.5465252255262278,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047626509174671214,
      "loss": 5.0573,
      "mean_token_accuracy": 0.20031104385852813,
      "num_tokens": 34055735.0,
      "step": 18515
    },
    {
      "entropy": 5.268128347396851,
      "epoch": 1.5469428666889409,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004762517814684187,
      "loss": 4.9988,
      "mean_token_accuracy": 0.1995690032839775,
      "num_tokens": 34065780.0,
      "step": 18520
    },
    {
      "entropy": 5.162655353546143,
      "epoch": 1.547360507851654,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047623846766697056,
      "loss": 4.9123,
      "mean_token_accuracy": 0.20614407509565352,
      "num_tokens": 34075314.0,
      "step": 18525
    },
    {
      "entropy": 5.194657039642334,
      "epoch": 1.5477781490143667,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047622515034260077,
      "loss": 5.0061,
      "mean_token_accuracy": 0.20638622343540192,
      "num_tokens": 34085475.0,
      "step": 18530
    },
    {
      "entropy": 5.2449596405029295,
      "epoch": 1.54819579017708,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004762118294955426,
      "loss": 4.9712,
      "mean_token_accuracy": 0.20112818628549575,
      "num_tokens": 34094758.0,
      "step": 18535
    },
    {
      "entropy": 5.1798192977905275,
      "epoch": 1.5486134313397928,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004761985051260291,
      "loss": 5.0009,
      "mean_token_accuracy": 0.20503217577934266,
      "num_tokens": 34103605.0,
      "step": 18540
    },
    {
      "entropy": 5.160132980346679,
      "epoch": 1.5490310725025058,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047618517723429366,
      "loss": 4.8505,
      "mean_token_accuracy": 0.2058885872364044,
      "num_tokens": 34112751.0,
      "step": 18545
    },
    {
      "entropy": 5.1587645530700685,
      "epoch": 1.5494487136652189,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047617184582056946,
      "loss": 4.9369,
      "mean_token_accuracy": 0.20514457374811174,
      "num_tokens": 34121814.0,
      "step": 18550
    },
    {
      "entropy": 5.099225378036499,
      "epoch": 1.549866354827932,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047615851088508995,
      "loss": 4.8681,
      "mean_token_accuracy": 0.21279157400131227,
      "num_tokens": 34130297.0,
      "step": 18555
    },
    {
      "entropy": 5.130818653106689,
      "epoch": 1.550283995990645,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047614517242808866,
      "loss": 4.8651,
      "mean_token_accuracy": 0.20808970034122468,
      "num_tokens": 34138525.0,
      "step": 18560
    },
    {
      "entropy": 5.05301833152771,
      "epoch": 1.5507016371533577,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00047613183044979885,
      "loss": 4.7911,
      "mean_token_accuracy": 0.2185426965355873,
      "num_tokens": 34147029.0,
      "step": 18565
    },
    {
      "entropy": 5.159122323989868,
      "epoch": 1.551119278316071,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047611848495045434,
      "loss": 4.9429,
      "mean_token_accuracy": 0.20499929487705232,
      "num_tokens": 34155617.0,
      "step": 18570
    },
    {
      "entropy": 5.225042819976807,
      "epoch": 1.5515369194787838,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047610513593028867,
      "loss": 4.9494,
      "mean_token_accuracy": 0.20593419671058655,
      "num_tokens": 34164693.0,
      "step": 18575
    },
    {
      "entropy": 5.126499462127685,
      "epoch": 1.5519545606414968,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004760917833895355,
      "loss": 4.9016,
      "mean_token_accuracy": 0.21016495674848557,
      "num_tokens": 34173394.0,
      "step": 18580
    },
    {
      "entropy": 5.104201078414917,
      "epoch": 1.5523722018042099,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004760784273284286,
      "loss": 4.8898,
      "mean_token_accuracy": 0.2145230770111084,
      "num_tokens": 34182522.0,
      "step": 18585
    },
    {
      "entropy": 5.158851480484008,
      "epoch": 1.5527898429669227,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004760650677472018,
      "loss": 4.9208,
      "mean_token_accuracy": 0.20717516541481018,
      "num_tokens": 34192456.0,
      "step": 18590
    },
    {
      "entropy": 5.284269666671753,
      "epoch": 1.553207484129636,
      "grad_norm": 1.34375,
      "learning_rate": 0.000476051704646089,
      "loss": 5.0602,
      "mean_token_accuracy": 0.19965837448835372,
      "num_tokens": 34202360.0,
      "step": 18595
    },
    {
      "entropy": 5.250815105438233,
      "epoch": 1.5536251252923488,
      "grad_norm": 1.21875,
      "learning_rate": 0.000476038338025324,
      "loss": 4.9353,
      "mean_token_accuracy": 0.20750900208950043,
      "num_tokens": 34211892.0,
      "step": 18600
    },
    {
      "entropy": 5.1277055740356445,
      "epoch": 1.5540427664550618,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047602496788514103,
      "loss": 4.9464,
      "mean_token_accuracy": 0.20691440403461456,
      "num_tokens": 34221776.0,
      "step": 18605
    },
    {
      "entropy": 5.094801139831543,
      "epoch": 1.5544604076177748,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004760115942257739,
      "loss": 4.8864,
      "mean_token_accuracy": 0.21093628108501433,
      "num_tokens": 34230564.0,
      "step": 18610
    },
    {
      "entropy": 5.067476463317871,
      "epoch": 1.5548780487804879,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047599821704745693,
      "loss": 4.8999,
      "mean_token_accuracy": 0.21348951756954193,
      "num_tokens": 34239871.0,
      "step": 18615
    },
    {
      "entropy": 5.141977691650391,
      "epoch": 1.5552956899432009,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004759848363504243,
      "loss": 4.9318,
      "mean_token_accuracy": 0.21370896250009536,
      "num_tokens": 34248881.0,
      "step": 18620
    },
    {
      "entropy": 5.2012323379516605,
      "epoch": 1.5557133311059137,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004759714521349101,
      "loss": 4.9058,
      "mean_token_accuracy": 0.2107441633939743,
      "num_tokens": 34257596.0,
      "step": 18625
    },
    {
      "entropy": 5.1171770095825195,
      "epoch": 1.556130972268627,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004759580644011487,
      "loss": 4.9189,
      "mean_token_accuracy": 0.20938018411397935,
      "num_tokens": 34267030.0,
      "step": 18630
    },
    {
      "entropy": 5.1882322311401365,
      "epoch": 1.5565486134313398,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004759446731493745,
      "loss": 4.9514,
      "mean_token_accuracy": 0.20286479890346526,
      "num_tokens": 34276269.0,
      "step": 18635
    },
    {
      "entropy": 5.131815814971924,
      "epoch": 1.5569662545940528,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000475931278379822,
      "loss": 4.883,
      "mean_token_accuracy": 0.20935932993888856,
      "num_tokens": 34285193.0,
      "step": 18640
    },
    {
      "entropy": 5.079673147201538,
      "epoch": 1.5573838957567658,
      "grad_norm": 1.25,
      "learning_rate": 0.0004759178800927255,
      "loss": 4.7947,
      "mean_token_accuracy": 0.21424604058265687,
      "num_tokens": 34294216.0,
      "step": 18645
    },
    {
      "entropy": 5.1956781387329105,
      "epoch": 1.5578015369194786,
      "grad_norm": 1.375,
      "learning_rate": 0.0004759044782883198,
      "loss": 5.023,
      "mean_token_accuracy": 0.2132480263710022,
      "num_tokens": 34302546.0,
      "step": 18650
    },
    {
      "entropy": 5.187273693084717,
      "epoch": 1.558219178082192,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004758910729668392,
      "loss": 4.8206,
      "mean_token_accuracy": 0.2160439148545265,
      "num_tokens": 34312176.0,
      "step": 18655
    },
    {
      "entropy": 5.236117506027222,
      "epoch": 1.5586368192449047,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004758776641285187,
      "loss": 4.9405,
      "mean_token_accuracy": 0.20605048686265945,
      "num_tokens": 34321836.0,
      "step": 18660
    },
    {
      "entropy": 5.116193008422852,
      "epoch": 1.5590544604076177,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004758642517735928,
      "loss": 4.9279,
      "mean_token_accuracy": 0.21284099817276,
      "num_tokens": 34330624.0,
      "step": 18665
    },
    {
      "entropy": 5.105957651138306,
      "epoch": 1.5594721015703308,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047585083590229643,
      "loss": 4.8421,
      "mean_token_accuracy": 0.21769220978021622,
      "num_tokens": 34339239.0,
      "step": 18670
    },
    {
      "entropy": 5.213536310195923,
      "epoch": 1.5598897427330438,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047583741651486443,
      "loss": 4.9603,
      "mean_token_accuracy": 0.20396847724914552,
      "num_tokens": 34348823.0,
      "step": 18675
    },
    {
      "entropy": 5.015321969985962,
      "epoch": 1.5603073838957568,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004758239936115317,
      "loss": 4.808,
      "mean_token_accuracy": 0.21196022927761077,
      "num_tokens": 34357844.0,
      "step": 18680
    },
    {
      "entropy": 5.142438077926636,
      "epoch": 1.5607250250584697,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047581056719253324,
      "loss": 4.8878,
      "mean_token_accuracy": 0.2076462358236313,
      "num_tokens": 34367055.0,
      "step": 18685
    },
    {
      "entropy": 5.273908615112305,
      "epoch": 1.561142666221183,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004757971372581041,
      "loss": 5.0689,
      "mean_token_accuracy": 0.20363930910825728,
      "num_tokens": 34377329.0,
      "step": 18690
    },
    {
      "entropy": 5.193539237976074,
      "epoch": 1.5615603073838957,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00047578370380847935,
      "loss": 5.052,
      "mean_token_accuracy": 0.1958830714225769,
      "num_tokens": 34386508.0,
      "step": 18695
    },
    {
      "entropy": 5.161758708953857,
      "epoch": 1.5619779485466088,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047577026684389413,
      "loss": 4.9221,
      "mean_token_accuracy": 0.20134143978357316,
      "num_tokens": 34396075.0,
      "step": 18700
    },
    {
      "entropy": 5.207879686355591,
      "epoch": 1.5623955897093218,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004757568263645838,
      "loss": 4.9789,
      "mean_token_accuracy": 0.20375854820013045,
      "num_tokens": 34405409.0,
      "step": 18705
    },
    {
      "entropy": 5.190208578109742,
      "epoch": 1.5628132308720346,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004757433823707835,
      "loss": 4.9454,
      "mean_token_accuracy": 0.2067614644765854,
      "num_tokens": 34415559.0,
      "step": 18710
    },
    {
      "entropy": 5.188376712799072,
      "epoch": 1.5632308720347479,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004757299348627287,
      "loss": 4.9466,
      "mean_token_accuracy": 0.21164121478796005,
      "num_tokens": 34424137.0,
      "step": 18715
    },
    {
      "entropy": 5.100105905532837,
      "epoch": 1.5636485131974607,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047571648384065477,
      "loss": 4.9518,
      "mean_token_accuracy": 0.20568589568138124,
      "num_tokens": 34433498.0,
      "step": 18720
    },
    {
      "entropy": 5.15032844543457,
      "epoch": 1.5640661543601737,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004757030293047972,
      "loss": 4.902,
      "mean_token_accuracy": 0.21556744277477263,
      "num_tokens": 34443218.0,
      "step": 18725
    },
    {
      "entropy": 5.135912036895752,
      "epoch": 1.5644837955228867,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004756895712553915,
      "loss": 4.9038,
      "mean_token_accuracy": 0.2080191746354103,
      "num_tokens": 34452981.0,
      "step": 18730
    },
    {
      "entropy": 5.120395231246948,
      "epoch": 1.5649014366855998,
      "grad_norm": 1.28125,
      "learning_rate": 0.00047567610969267326,
      "loss": 4.9353,
      "mean_token_accuracy": 0.2004684552550316,
      "num_tokens": 34462763.0,
      "step": 18735
    },
    {
      "entropy": 5.122578716278076,
      "epoch": 1.5653190778483128,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004756626446168782,
      "loss": 4.834,
      "mean_token_accuracy": 0.21418932527303697,
      "num_tokens": 34471841.0,
      "step": 18740
    },
    {
      "entropy": 5.251544046401977,
      "epoch": 1.5657367190110256,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047564917602824203,
      "loss": 5.0496,
      "mean_token_accuracy": 0.20416231751441954,
      "num_tokens": 34482280.0,
      "step": 18745
    },
    {
      "entropy": 5.296657514572144,
      "epoch": 1.5661543601737389,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004756357039270004,
      "loss": 5.0979,
      "mean_token_accuracy": 0.20085369646549225,
      "num_tokens": 34492774.0,
      "step": 18750
    },
    {
      "entropy": 5.155768394470215,
      "epoch": 1.5665720013364517,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047562222831338933,
      "loss": 4.8418,
      "mean_token_accuracy": 0.21169922202825547,
      "num_tokens": 34501626.0,
      "step": 18755
    },
    {
      "entropy": 5.126046752929687,
      "epoch": 1.5669896424991647,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047560874918764473,
      "loss": 4.9459,
      "mean_token_accuracy": 0.20495051592588426,
      "num_tokens": 34510264.0,
      "step": 18760
    },
    {
      "entropy": 5.230197620391846,
      "epoch": 1.5674072836618778,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047559526655000244,
      "loss": 4.9956,
      "mean_token_accuracy": 0.19665072113275528,
      "num_tokens": 34520540.0,
      "step": 18765
    },
    {
      "entropy": 5.174895763397217,
      "epoch": 1.5678249248245906,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004755817804006986,
      "loss": 4.9426,
      "mean_token_accuracy": 0.21024518460035324,
      "num_tokens": 34529669.0,
      "step": 18770
    },
    {
      "entropy": 5.151412487030029,
      "epoch": 1.5682425659873038,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004755682907399692,
      "loss": 4.93,
      "mean_token_accuracy": 0.20811665505170823,
      "num_tokens": 34538802.0,
      "step": 18775
    },
    {
      "entropy": 5.159359312057495,
      "epoch": 1.5686602071500166,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00047555479756805055,
      "loss": 4.998,
      "mean_token_accuracy": 0.2042615681886673,
      "num_tokens": 34547583.0,
      "step": 18780
    },
    {
      "entropy": 5.164404392242432,
      "epoch": 1.5690778483127297,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004755413008851786,
      "loss": 4.8779,
      "mean_token_accuracy": 0.21310614347457885,
      "num_tokens": 34556045.0,
      "step": 18785
    },
    {
      "entropy": 5.120488834381104,
      "epoch": 1.5694954894754427,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004755278006915899,
      "loss": 4.7752,
      "mean_token_accuracy": 0.21924481093883513,
      "num_tokens": 34564910.0,
      "step": 18790
    },
    {
      "entropy": 5.097643852233887,
      "epoch": 1.5699131306381557,
      "grad_norm": 1.375,
      "learning_rate": 0.00047551429698752064,
      "loss": 5.005,
      "mean_token_accuracy": 0.2022622600197792,
      "num_tokens": 34573274.0,
      "step": 18795
    },
    {
      "entropy": 5.13235387802124,
      "epoch": 1.5703307718008688,
      "grad_norm": 2.25,
      "learning_rate": 0.00047550078977320724,
      "loss": 4.8834,
      "mean_token_accuracy": 0.21242711395025254,
      "num_tokens": 34583015.0,
      "step": 18800
    },
    {
      "entropy": 5.254848527908325,
      "epoch": 1.5707484129635816,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004754872790488862,
      "loss": 5.0049,
      "mean_token_accuracy": 0.20000348389148712,
      "num_tokens": 34592311.0,
      "step": 18805
    },
    {
      "entropy": 5.199438142776489,
      "epoch": 1.5711660541262948,
      "grad_norm": 1.234375,
      "learning_rate": 0.000475473764814794,
      "loss": 4.8883,
      "mean_token_accuracy": 0.21494589298963546,
      "num_tokens": 34600806.0,
      "step": 18810
    },
    {
      "entropy": 5.119493770599365,
      "epoch": 1.5715836952890077,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047546024707116727,
      "loss": 4.8552,
      "mean_token_accuracy": 0.20589278489351273,
      "num_tokens": 34609830.0,
      "step": 18815
    },
    {
      "entropy": 5.1449951171875,
      "epoch": 1.5720013364517207,
      "grad_norm": 1.34375,
      "learning_rate": 0.00047544672581824256,
      "loss": 4.8885,
      "mean_token_accuracy": 0.2051151141524315,
      "num_tokens": 34618870.0,
      "step": 18820
    },
    {
      "entropy": 5.07779598236084,
      "epoch": 1.5724189776144337,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004754332010562567,
      "loss": 4.8262,
      "mean_token_accuracy": 0.21625585705041886,
      "num_tokens": 34628349.0,
      "step": 18825
    },
    {
      "entropy": 5.151827335357666,
      "epoch": 1.5728366187771465,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004754196727854463,
      "loss": 4.9613,
      "mean_token_accuracy": 0.20374633222818375,
      "num_tokens": 34638492.0,
      "step": 18830
    },
    {
      "entropy": 5.23334584236145,
      "epoch": 1.5732542599398598,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047540614100604836,
      "loss": 5.0328,
      "mean_token_accuracy": 0.208150514960289,
      "num_tokens": 34649084.0,
      "step": 18835
    },
    {
      "entropy": 5.178091526031494,
      "epoch": 1.5736719011025726,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004753926057182997,
      "loss": 4.9124,
      "mean_token_accuracy": 0.21244193166494368,
      "num_tokens": 34658392.0,
      "step": 18840
    },
    {
      "entropy": 5.098398208618164,
      "epoch": 1.5740895422652856,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047537906692243716,
      "loss": 4.9628,
      "mean_token_accuracy": 0.2002706378698349,
      "num_tokens": 34667727.0,
      "step": 18845
    },
    {
      "entropy": 5.131758546829223,
      "epoch": 1.5745071834279987,
      "grad_norm": 1.21875,
      "learning_rate": 0.000475365524618698,
      "loss": 4.912,
      "mean_token_accuracy": 0.20995223820209502,
      "num_tokens": 34676260.0,
      "step": 18850
    },
    {
      "entropy": 5.283615207672119,
      "epoch": 1.5749248245907117,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004753519788073191,
      "loss": 5.0467,
      "mean_token_accuracy": 0.19865862131118775,
      "num_tokens": 34685145.0,
      "step": 18855
    },
    {
      "entropy": 5.120633840560913,
      "epoch": 1.5753424657534247,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004753384294885376,
      "loss": 4.823,
      "mean_token_accuracy": 0.2208677649497986,
      "num_tokens": 34694035.0,
      "step": 18860
    },
    {
      "entropy": 5.109022712707519,
      "epoch": 1.5757601069161375,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004753248766625908,
      "loss": 4.9137,
      "mean_token_accuracy": 0.20451402217149733,
      "num_tokens": 34702634.0,
      "step": 18865
    },
    {
      "entropy": 5.121854209899903,
      "epoch": 1.5761777480788508,
      "grad_norm": 1.25,
      "learning_rate": 0.0004753113203297159,
      "loss": 4.9185,
      "mean_token_accuracy": 0.20061037838459014,
      "num_tokens": 34711925.0,
      "step": 18870
    },
    {
      "entropy": 5.060426950454712,
      "epoch": 1.5765953892415636,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004752977604901502,
      "loss": 4.8438,
      "mean_token_accuracy": 0.22011109739542006,
      "num_tokens": 34720131.0,
      "step": 18875
    },
    {
      "entropy": 5.116326427459716,
      "epoch": 1.5770130304042767,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047528419714413107,
      "loss": 4.9714,
      "mean_token_accuracy": 0.2072340115904808,
      "num_tokens": 34728869.0,
      "step": 18880
    },
    {
      "entropy": 5.1223876953125,
      "epoch": 1.5774306715669897,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047527063029189606,
      "loss": 4.8148,
      "mean_token_accuracy": 0.22200645208358766,
      "num_tokens": 34737343.0,
      "step": 18885
    },
    {
      "entropy": 5.216577196121216,
      "epoch": 1.5778483127297025,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004752570599336825,
      "loss": 5.0066,
      "mean_token_accuracy": 0.2012176424264908,
      "num_tokens": 34746752.0,
      "step": 18890
    },
    {
      "entropy": 5.219081211090088,
      "epoch": 1.5782659538924158,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047524348606972817,
      "loss": 4.9465,
      "mean_token_accuracy": 0.20830136984586717,
      "num_tokens": 34755672.0,
      "step": 18895
    },
    {
      "entropy": 5.162012767791748,
      "epoch": 1.5786835950551286,
      "grad_norm": 1.390625,
      "learning_rate": 0.00047522990870027055,
      "loss": 4.9599,
      "mean_token_accuracy": 0.20078527331352233,
      "num_tokens": 34764558.0,
      "step": 18900
    },
    {
      "entropy": 5.129535675048828,
      "epoch": 1.5791012362178416,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004752163278255473,
      "loss": 4.878,
      "mean_token_accuracy": 0.20477416813373567,
      "num_tokens": 34773663.0,
      "step": 18905
    },
    {
      "entropy": 5.041424751281738,
      "epoch": 1.5795188773805546,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004752027434457963,
      "loss": 4.8227,
      "mean_token_accuracy": 0.2136944428086281,
      "num_tokens": 34782297.0,
      "step": 18910
    },
    {
      "entropy": 5.278899192810059,
      "epoch": 1.5799365185432677,
      "grad_norm": 1.25,
      "learning_rate": 0.0004751891555612552,
      "loss": 5.1059,
      "mean_token_accuracy": 0.19408555328845978,
      "num_tokens": 34791734.0,
      "step": 18915
    },
    {
      "entropy": 5.166095542907715,
      "epoch": 1.5803541597059807,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004751755641721621,
      "loss": 4.9149,
      "mean_token_accuracy": 0.2111927330493927,
      "num_tokens": 34801187.0,
      "step": 18920
    },
    {
      "entropy": 5.143557500839234,
      "epoch": 1.5807718008686935,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047516196927875474,
      "loss": 4.9176,
      "mean_token_accuracy": 0.20845446288585662,
      "num_tokens": 34810084.0,
      "step": 18925
    },
    {
      "entropy": 5.32639331817627,
      "epoch": 1.5811894420314068,
      "grad_norm": 1.265625,
      "learning_rate": 0.00047514837088127117,
      "loss": 5.0178,
      "mean_token_accuracy": 0.20774588584899903,
      "num_tokens": 34819160.0,
      "step": 18930
    },
    {
      "entropy": 5.093873023986816,
      "epoch": 1.5816070831941196,
      "grad_norm": 1.375,
      "learning_rate": 0.0004751347689799494,
      "loss": 4.8204,
      "mean_token_accuracy": 0.22372463196516038,
      "num_tokens": 34828272.0,
      "step": 18935
    },
    {
      "entropy": 5.03593602180481,
      "epoch": 1.5820247243568326,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047512116357502766,
      "loss": 4.8577,
      "mean_token_accuracy": 0.20925870090723037,
      "num_tokens": 34838030.0,
      "step": 18940
    },
    {
      "entropy": 5.1505204200744625,
      "epoch": 1.5824423655195456,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047510755466674407,
      "loss": 4.9561,
      "mean_token_accuracy": 0.20542921870946884,
      "num_tokens": 34846489.0,
      "step": 18945
    },
    {
      "entropy": 5.183438920974732,
      "epoch": 1.5828600066822585,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047509394225533685,
      "loss": 4.9058,
      "mean_token_accuracy": 0.21159907430410385,
      "num_tokens": 34855926.0,
      "step": 18950
    },
    {
      "entropy": 5.161541986465454,
      "epoch": 1.5832776478449717,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004750803263410443,
      "loss": 4.9041,
      "mean_token_accuracy": 0.2065946117043495,
      "num_tokens": 34865158.0,
      "step": 18955
    },
    {
      "entropy": 5.145259618759155,
      "epoch": 1.5836952890076845,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004750667069241048,
      "loss": 4.9692,
      "mean_token_accuracy": 0.20303819328546524,
      "num_tokens": 34875429.0,
      "step": 18960
    },
    {
      "entropy": 5.174824953079224,
      "epoch": 1.5841129301703976,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004750530840047568,
      "loss": 4.9583,
      "mean_token_accuracy": 0.20479912012815477,
      "num_tokens": 34885663.0,
      "step": 18965
    },
    {
      "entropy": 5.1419189453125,
      "epoch": 1.5845305713331106,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047503945758323873,
      "loss": 4.8627,
      "mean_token_accuracy": 0.2139056697487831,
      "num_tokens": 34894976.0,
      "step": 18970
    },
    {
      "entropy": 5.195501804351807,
      "epoch": 1.5849482124958236,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004750258276597892,
      "loss": 4.8893,
      "mean_token_accuracy": 0.20480023920536042,
      "num_tokens": 34905630.0,
      "step": 18975
    },
    {
      "entropy": 5.104459667205811,
      "epoch": 1.5853658536585367,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047501219423464673,
      "loss": 4.8489,
      "mean_token_accuracy": 0.21074771881103516,
      "num_tokens": 34914906.0,
      "step": 18980
    },
    {
      "entropy": 5.059500455856323,
      "epoch": 1.5857834948212495,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004749985573080501,
      "loss": 4.8232,
      "mean_token_accuracy": 0.21005662977695466,
      "num_tokens": 34923274.0,
      "step": 18985
    },
    {
      "entropy": 5.141012668609619,
      "epoch": 1.5862011359839627,
      "grad_norm": 1.171875,
      "learning_rate": 0.000474984916880238,
      "loss": 4.8838,
      "mean_token_accuracy": 0.21457558125257492,
      "num_tokens": 34932085.0,
      "step": 18990
    },
    {
      "entropy": 5.102866458892822,
      "epoch": 1.5866187771466755,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004749712729514491,
      "loss": 4.9627,
      "mean_token_accuracy": 0.2014758825302124,
      "num_tokens": 34940871.0,
      "step": 18995
    },
    {
      "entropy": 5.211615085601807,
      "epoch": 1.5870364183093886,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004749576255219225,
      "loss": 5.0169,
      "mean_token_accuracy": 0.2047075867652893,
      "num_tokens": 34950703.0,
      "step": 19000
    },
    {
      "entropy": 5.252520418167114,
      "epoch": 1.5874540594721016,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004749439745918969,
      "loss": 4.9724,
      "mean_token_accuracy": 0.2040489822626114,
      "num_tokens": 34959988.0,
      "step": 19005
    },
    {
      "entropy": 5.1748981952667235,
      "epoch": 1.5878717006348144,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047493032016161134,
      "loss": 4.9685,
      "mean_token_accuracy": 0.20617054998874665,
      "num_tokens": 34968608.0,
      "step": 19010
    },
    {
      "entropy": 5.235073232650757,
      "epoch": 1.5882893417975277,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004749166622313049,
      "loss": 4.9849,
      "mean_token_accuracy": 0.1964227259159088,
      "num_tokens": 34977200.0,
      "step": 19015
    },
    {
      "entropy": 5.194996643066406,
      "epoch": 1.5887069829602405,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047490300080121667,
      "loss": 4.8886,
      "mean_token_accuracy": 0.21451182812452316,
      "num_tokens": 34986403.0,
      "step": 19020
    },
    {
      "entropy": 5.111818742752075,
      "epoch": 1.5891246241229535,
      "grad_norm": 1.3125,
      "learning_rate": 0.00047488933587158585,
      "loss": 5.0021,
      "mean_token_accuracy": 0.20476340055465697,
      "num_tokens": 34995672.0,
      "step": 19025
    },
    {
      "entropy": 5.144380331039429,
      "epoch": 1.5895422652856666,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004748756674426514,
      "loss": 5.0363,
      "mean_token_accuracy": 0.20398048013448716,
      "num_tokens": 35004636.0,
      "step": 19030
    },
    {
      "entropy": 5.191207647323608,
      "epoch": 1.5899599064483796,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000474861995514653,
      "loss": 4.8296,
      "mean_token_accuracy": 0.21595600992441177,
      "num_tokens": 35014023.0,
      "step": 19035
    },
    {
      "entropy": 5.2480885028839115,
      "epoch": 1.5903775476110926,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004748483200878297,
      "loss": 5.0354,
      "mean_token_accuracy": 0.20618441700935364,
      "num_tokens": 35023762.0,
      "step": 19040
    },
    {
      "entropy": 5.157561349868774,
      "epoch": 1.5907951887738054,
      "grad_norm": 1.234375,
      "learning_rate": 0.000474834641162421,
      "loss": 4.8793,
      "mean_token_accuracy": 0.2077969193458557,
      "num_tokens": 35032485.0,
      "step": 19045
    },
    {
      "entropy": 5.185596179962158,
      "epoch": 1.5912128299365187,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00047482095873866635,
      "loss": 4.9634,
      "mean_token_accuracy": 0.20305103659629822,
      "num_tokens": 35041416.0,
      "step": 19050
    },
    {
      "entropy": 5.223783874511719,
      "epoch": 1.5916304710992315,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004748072728168053,
      "loss": 4.8952,
      "mean_token_accuracy": 0.2061702311038971,
      "num_tokens": 35051223.0,
      "step": 19055
    },
    {
      "entropy": 5.126621198654175,
      "epoch": 1.5920481122619445,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004747935833970775,
      "loss": 4.9613,
      "mean_token_accuracy": 0.2073996290564537,
      "num_tokens": 35060935.0,
      "step": 19060
    },
    {
      "entropy": 5.098575782775879,
      "epoch": 1.5924657534246576,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004747798904797225,
      "loss": 4.8517,
      "mean_token_accuracy": 0.21550240516662597,
      "num_tokens": 35070249.0,
      "step": 19065
    },
    {
      "entropy": 5.12057671546936,
      "epoch": 1.5928833945873704,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047476619406497996,
      "loss": 4.8624,
      "mean_token_accuracy": 0.20905472934246064,
      "num_tokens": 35079040.0,
      "step": 19070
    },
    {
      "entropy": 5.0515203952789305,
      "epoch": 1.5933010357500836,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004747524941530898,
      "loss": 4.8216,
      "mean_token_accuracy": 0.21593879759311677,
      "num_tokens": 35087451.0,
      "step": 19075
    },
    {
      "entropy": 5.213313245773316,
      "epoch": 1.5937186769127965,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004747387907442918,
      "loss": 4.9498,
      "mean_token_accuracy": 0.2105334609746933,
      "num_tokens": 35096144.0,
      "step": 19080
    },
    {
      "entropy": 5.278976917266846,
      "epoch": 1.5941363180755095,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047472508383882586,
      "loss": 4.95,
      "mean_token_accuracy": 0.20482726246118546,
      "num_tokens": 35105547.0,
      "step": 19085
    },
    {
      "entropy": 5.146666526794434,
      "epoch": 1.5945539592382225,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004747113734369319,
      "loss": 4.9008,
      "mean_token_accuracy": 0.2093447342514992,
      "num_tokens": 35114164.0,
      "step": 19090
    },
    {
      "entropy": 5.101911640167236,
      "epoch": 1.5949716004009356,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004746976595388499,
      "loss": 4.9511,
      "mean_token_accuracy": 0.20196350663900375,
      "num_tokens": 35125184.0,
      "step": 19095
    },
    {
      "entropy": 5.125514173507691,
      "epoch": 1.5953892415636486,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047468394214482,
      "loss": 4.8317,
      "mean_token_accuracy": 0.21543429642915726,
      "num_tokens": 35134207.0,
      "step": 19100
    },
    {
      "entropy": 5.1058347702026365,
      "epoch": 1.5958068827263614,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004746702212550824,
      "loss": 4.9027,
      "mean_token_accuracy": 0.21508298367261885,
      "num_tokens": 35143651.0,
      "step": 19105
    },
    {
      "entropy": 5.12345142364502,
      "epoch": 1.5962245238890747,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047465649686987717,
      "loss": 4.9478,
      "mean_token_accuracy": 0.20518188625574113,
      "num_tokens": 35152907.0,
      "step": 19110
    },
    {
      "entropy": 5.144792222976685,
      "epoch": 1.5966421650517875,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004746427689894447,
      "loss": 4.8378,
      "mean_token_accuracy": 0.2187953770160675,
      "num_tokens": 35161927.0,
      "step": 19115
    },
    {
      "entropy": 5.163080739974975,
      "epoch": 1.5970598062145005,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004746290376140252,
      "loss": 4.9352,
      "mean_token_accuracy": 0.20987556576728822,
      "num_tokens": 35171303.0,
      "step": 19120
    },
    {
      "entropy": 5.238228750228882,
      "epoch": 1.5974774473772135,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004746153027438592,
      "loss": 5.0056,
      "mean_token_accuracy": 0.19962104856967927,
      "num_tokens": 35181160.0,
      "step": 19125
    },
    {
      "entropy": 5.203304481506348,
      "epoch": 1.5978950885399263,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004746015643791869,
      "loss": 4.9739,
      "mean_token_accuracy": 0.200551538169384,
      "num_tokens": 35190335.0,
      "step": 19130
    },
    {
      "entropy": 5.21612138748169,
      "epoch": 1.5983127297026396,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000474587822520249,
      "loss": 4.9838,
      "mean_token_accuracy": 0.20706733465194702,
      "num_tokens": 35200036.0,
      "step": 19135
    },
    {
      "entropy": 5.2018249988555905,
      "epoch": 1.5987303708653524,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004745740771672861,
      "loss": 4.8935,
      "mean_token_accuracy": 0.21380551308393478,
      "num_tokens": 35208203.0,
      "step": 19140
    },
    {
      "entropy": 5.0872578620910645,
      "epoch": 1.5991480120280654,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004745603283205387,
      "loss": 4.8365,
      "mean_token_accuracy": 0.21652160584926605,
      "num_tokens": 35216963.0,
      "step": 19145
    },
    {
      "entropy": 5.187752103805542,
      "epoch": 1.5995656531907785,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047454657598024755,
      "loss": 4.9153,
      "mean_token_accuracy": 0.2036829635500908,
      "num_tokens": 35226170.0,
      "step": 19150
    },
    {
      "entropy": 5.037940454483032,
      "epoch": 1.5999832943534915,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047453282014665334,
      "loss": 4.8017,
      "mean_token_accuracy": 0.21714409440755844,
      "num_tokens": 35234383.0,
      "step": 19155
    },
    {
      "entropy": 5.064695024490357,
      "epoch": 1.6004009355162045,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047451906081999695,
      "loss": 4.8204,
      "mean_token_accuracy": 0.21620627045631408,
      "num_tokens": 35242771.0,
      "step": 19160
    },
    {
      "entropy": 5.160063600540161,
      "epoch": 1.6008185766789174,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004745052980005193,
      "loss": 4.8811,
      "mean_token_accuracy": 0.21993630081415178,
      "num_tokens": 35251333.0,
      "step": 19165
    },
    {
      "entropy": 5.176940155029297,
      "epoch": 1.6012362178416306,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004744915316884612,
      "loss": 4.9502,
      "mean_token_accuracy": 0.19951265305280685,
      "num_tokens": 35260423.0,
      "step": 19170
    },
    {
      "entropy": 5.135950565338135,
      "epoch": 1.6016538590043434,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004744777618840637,
      "loss": 4.8693,
      "mean_token_accuracy": 0.2162060633301735,
      "num_tokens": 35268838.0,
      "step": 19175
    },
    {
      "entropy": 5.174797868728637,
      "epoch": 1.6020715001670565,
      "grad_norm": 1.125,
      "learning_rate": 0.000474463988587568,
      "loss": 5.0096,
      "mean_token_accuracy": 0.2051709085702896,
      "num_tokens": 35279395.0,
      "step": 19180
    },
    {
      "entropy": 5.2045663356781,
      "epoch": 1.6024891413297695,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047445021179921495,
      "loss": 4.9402,
      "mean_token_accuracy": 0.20374660491943358,
      "num_tokens": 35289637.0,
      "step": 19185
    },
    {
      "entropy": 5.131990051269531,
      "epoch": 1.6029067824924823,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004744364315192459,
      "loss": 4.9105,
      "mean_token_accuracy": 0.21145320534706116,
      "num_tokens": 35299813.0,
      "step": 19190
    },
    {
      "entropy": 5.142439603805542,
      "epoch": 1.6033244236551956,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047442264774790203,
      "loss": 4.9332,
      "mean_token_accuracy": 0.20990714877843858,
      "num_tokens": 35308162.0,
      "step": 19195
    },
    {
      "entropy": 5.076178359985351,
      "epoch": 1.6037420648179084,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004744088604854248,
      "loss": 4.8966,
      "mean_token_accuracy": 0.20883092433214187,
      "num_tokens": 35317232.0,
      "step": 19200
    },
    {
      "entropy": 5.106534385681153,
      "epoch": 1.6041597059806214,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047439506973205527,
      "loss": 4.8746,
      "mean_token_accuracy": 0.21035057157278061,
      "num_tokens": 35326011.0,
      "step": 19205
    },
    {
      "entropy": 5.058756256103516,
      "epoch": 1.6045773471433344,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004743812754880351,
      "loss": 4.8006,
      "mean_token_accuracy": 0.2179470866918564,
      "num_tokens": 35333935.0,
      "step": 19210
    },
    {
      "entropy": 5.166488313674927,
      "epoch": 1.6049949883060475,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004743674777536057,
      "loss": 4.9154,
      "mean_token_accuracy": 0.20577628910541534,
      "num_tokens": 35343512.0,
      "step": 19215
    },
    {
      "entropy": 5.163171815872192,
      "epoch": 1.6054126294687605,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004743536765290087,
      "loss": 4.9264,
      "mean_token_accuracy": 0.20465486347675324,
      "num_tokens": 35353276.0,
      "step": 19220
    },
    {
      "entropy": 5.164973306655884,
      "epoch": 1.6058302706314733,
      "grad_norm": 1.25,
      "learning_rate": 0.00047433987181448546,
      "loss": 4.9148,
      "mean_token_accuracy": 0.21246787607669831,
      "num_tokens": 35362101.0,
      "step": 19225
    },
    {
      "entropy": 5.085865497589111,
      "epoch": 1.6062479117941866,
      "grad_norm": 1.25,
      "learning_rate": 0.0004743260636102779,
      "loss": 4.8898,
      "mean_token_accuracy": 0.20804720669984816,
      "num_tokens": 35370998.0,
      "step": 19230
    },
    {
      "entropy": 5.228740787506103,
      "epoch": 1.6066655529568994,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047431225191662776,
      "loss": 5.0068,
      "mean_token_accuracy": 0.20112838745117187,
      "num_tokens": 35380347.0,
      "step": 19235
    },
    {
      "entropy": 5.196119117736816,
      "epoch": 1.6070831941196124,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047429843673377664,
      "loss": 4.9001,
      "mean_token_accuracy": 0.2134018823504448,
      "num_tokens": 35388586.0,
      "step": 19240
    },
    {
      "entropy": 5.171203088760376,
      "epoch": 1.6075008352823255,
      "grad_norm": 1.34375,
      "learning_rate": 0.00047428461806196646,
      "loss": 4.9832,
      "mean_token_accuracy": 0.20613851845264436,
      "num_tokens": 35397649.0,
      "step": 19245
    },
    {
      "entropy": 5.106731224060058,
      "epoch": 1.6079184764450383,
      "grad_norm": 1.25,
      "learning_rate": 0.0004742707959014393,
      "loss": 4.9064,
      "mean_token_accuracy": 0.21025595515966417,
      "num_tokens": 35405790.0,
      "step": 19250
    },
    {
      "entropy": 5.1246847152709964,
      "epoch": 1.6083361176077515,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047425697025243693,
      "loss": 4.9329,
      "mean_token_accuracy": 0.20134546160697936,
      "num_tokens": 35414389.0,
      "step": 19255
    },
    {
      "entropy": 5.220750761032105,
      "epoch": 1.6087537587704643,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047424314111520134,
      "loss": 4.9366,
      "mean_token_accuracy": 0.2108313575387001,
      "num_tokens": 35424909.0,
      "step": 19260
    },
    {
      "entropy": 5.159171533584595,
      "epoch": 1.6091713999331774,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004742293084899749,
      "loss": 4.9007,
      "mean_token_accuracy": 0.21125598400831222,
      "num_tokens": 35434054.0,
      "step": 19265
    },
    {
      "entropy": 5.124175071716309,
      "epoch": 1.6095890410958904,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004742154723769995,
      "loss": 4.9093,
      "mean_token_accuracy": 0.21118346005678176,
      "num_tokens": 35443722.0,
      "step": 19270
    },
    {
      "entropy": 5.225421667098999,
      "epoch": 1.6100066822586034,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004742016327765175,
      "loss": 4.9384,
      "mean_token_accuracy": 0.2058065801858902,
      "num_tokens": 35453067.0,
      "step": 19275
    },
    {
      "entropy": 5.069595050811768,
      "epoch": 1.6104243234213165,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004741877896887711,
      "loss": 4.8191,
      "mean_token_accuracy": 0.21194617152214051,
      "num_tokens": 35462425.0,
      "step": 19280
    },
    {
      "entropy": 5.099846363067627,
      "epoch": 1.6108419645840293,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004741739431140027,
      "loss": 4.8658,
      "mean_token_accuracy": 0.21220189332962036,
      "num_tokens": 35472037.0,
      "step": 19285
    },
    {
      "entropy": 5.18293948173523,
      "epoch": 1.6112596057467425,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047416009305245474,
      "loss": 4.9206,
      "mean_token_accuracy": 0.20941727608442307,
      "num_tokens": 35481411.0,
      "step": 19290
    },
    {
      "entropy": 5.076171112060547,
      "epoch": 1.6116772469094554,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004741462395043695,
      "loss": 4.8328,
      "mean_token_accuracy": 0.22138451933860778,
      "num_tokens": 35490019.0,
      "step": 19295
    },
    {
      "entropy": 5.10058364868164,
      "epoch": 1.6120948880721684,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047413238246998975,
      "loss": 4.9347,
      "mean_token_accuracy": 0.21507761776447296,
      "num_tokens": 35499229.0,
      "step": 19300
    },
    {
      "entropy": 5.147928524017334,
      "epoch": 1.6125125292348814,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004741185219495579,
      "loss": 4.8372,
      "mean_token_accuracy": 0.2119302600622177,
      "num_tokens": 35508473.0,
      "step": 19305
    },
    {
      "entropy": 5.145115184783935,
      "epoch": 1.6129301703975942,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004741046579433166,
      "loss": 4.9261,
      "mean_token_accuracy": 0.21186739802360535,
      "num_tokens": 35517887.0,
      "step": 19310
    },
    {
      "entropy": 5.161832094192505,
      "epoch": 1.6133478115603075,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047409079045150865,
      "loss": 4.9037,
      "mean_token_accuracy": 0.21440439671278,
      "num_tokens": 35527470.0,
      "step": 19315
    },
    {
      "entropy": 5.126580238342285,
      "epoch": 1.6137654527230203,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047407691947437677,
      "loss": 4.9634,
      "mean_token_accuracy": 0.2084278866648674,
      "num_tokens": 35537670.0,
      "step": 19320
    },
    {
      "entropy": 5.09269151687622,
      "epoch": 1.6141830938857333,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047406304501216375,
      "loss": 4.8944,
      "mean_token_accuracy": 0.21244882196187972,
      "num_tokens": 35546903.0,
      "step": 19325
    },
    {
      "entropy": 5.166776323318482,
      "epoch": 1.6146007350484464,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047404916706511255,
      "loss": 4.896,
      "mean_token_accuracy": 0.20738958865404128,
      "num_tokens": 35556198.0,
      "step": 19330
    },
    {
      "entropy": 5.2503913879394535,
      "epoch": 1.6150183762111594,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00047403528563346603,
      "loss": 5.0353,
      "mean_token_accuracy": 0.19828736484050752,
      "num_tokens": 35566381.0,
      "step": 19335
    },
    {
      "entropy": 5.10810866355896,
      "epoch": 1.6154360173738724,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047402140071746743,
      "loss": 4.8783,
      "mean_token_accuracy": 0.2202883094549179,
      "num_tokens": 35575692.0,
      "step": 19340
    },
    {
      "entropy": 5.059160041809082,
      "epoch": 1.6158536585365852,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047400751231735946,
      "loss": 4.8967,
      "mean_token_accuracy": 0.20792653113603593,
      "num_tokens": 35584783.0,
      "step": 19345
    },
    {
      "entropy": 5.163923692703247,
      "epoch": 1.6162712996992985,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047399362043338547,
      "loss": 4.9452,
      "mean_token_accuracy": 0.20440325736999512,
      "num_tokens": 35595410.0,
      "step": 19350
    },
    {
      "entropy": 5.184153652191162,
      "epoch": 1.6166889408620113,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004739797250657887,
      "loss": 4.8427,
      "mean_token_accuracy": 0.2092247039079666,
      "num_tokens": 35604950.0,
      "step": 19355
    },
    {
      "entropy": 5.14867377281189,
      "epoch": 1.6171065820247243,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047396582621481224,
      "loss": 4.9353,
      "mean_token_accuracy": 0.19776372760534286,
      "num_tokens": 35614227.0,
      "step": 19360
    },
    {
      "entropy": 5.082727527618408,
      "epoch": 1.6175242231874374,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047395192388069955,
      "loss": 4.9507,
      "mean_token_accuracy": 0.20421325415372849,
      "num_tokens": 35623349.0,
      "step": 19365
    },
    {
      "entropy": 5.072706794738769,
      "epoch": 1.6179418643501502,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047393801806369396,
      "loss": 4.7522,
      "mean_token_accuracy": 0.22574374675750733,
      "num_tokens": 35632169.0,
      "step": 19370
    },
    {
      "entropy": 5.182081890106201,
      "epoch": 1.6183595055128634,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047392410876403887,
      "loss": 5.0117,
      "mean_token_accuracy": 0.20268008410930632,
      "num_tokens": 35642121.0,
      "step": 19375
    },
    {
      "entropy": 5.248095273971558,
      "epoch": 1.6187771466755763,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047391019598197785,
      "loss": 4.9563,
      "mean_token_accuracy": 0.2052820935845375,
      "num_tokens": 35651771.0,
      "step": 19380
    },
    {
      "entropy": 5.1404547691345215,
      "epoch": 1.6191947878382893,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004738962797177544,
      "loss": 4.9166,
      "mean_token_accuracy": 0.20583274960517883,
      "num_tokens": 35661422.0,
      "step": 19385
    },
    {
      "entropy": 5.161347103118897,
      "epoch": 1.6196124290010023,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004738823599716121,
      "loss": 5.0124,
      "mean_token_accuracy": 0.20035074055194854,
      "num_tokens": 35672245.0,
      "step": 19390
    },
    {
      "entropy": 5.139031505584716,
      "epoch": 1.6200300701637154,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047386843674379484,
      "loss": 4.9551,
      "mean_token_accuracy": 0.20096612125635147,
      "num_tokens": 35681690.0,
      "step": 19395
    },
    {
      "entropy": 5.135376691818237,
      "epoch": 1.6204477113264284,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004738545100345461,
      "loss": 4.8091,
      "mean_token_accuracy": 0.22142912447452545,
      "num_tokens": 35690265.0,
      "step": 19400
    },
    {
      "entropy": 5.164811658859253,
      "epoch": 1.6208653524891412,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047384057984410985,
      "loss": 4.9641,
      "mean_token_accuracy": 0.20353262275457382,
      "num_tokens": 35699768.0,
      "step": 19405
    },
    {
      "entropy": 5.116515111923218,
      "epoch": 1.6212829936518545,
      "grad_norm": 1.25,
      "learning_rate": 0.00047382664617272985,
      "loss": 4.9082,
      "mean_token_accuracy": 0.21121852546930314,
      "num_tokens": 35708785.0,
      "step": 19410
    },
    {
      "entropy": 5.151244306564331,
      "epoch": 1.6217006348145673,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047381270902065013,
      "loss": 4.9586,
      "mean_token_accuracy": 0.2023837685585022,
      "num_tokens": 35719224.0,
      "step": 19415
    },
    {
      "entropy": 5.143857383728028,
      "epoch": 1.6221182759772803,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047379876838811463,
      "loss": 4.9601,
      "mean_token_accuracy": 0.20105476081371307,
      "num_tokens": 35727830.0,
      "step": 19420
    },
    {
      "entropy": 5.147453308105469,
      "epoch": 1.6225359171399933,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004737848242753674,
      "loss": 4.9397,
      "mean_token_accuracy": 0.2052108332514763,
      "num_tokens": 35736818.0,
      "step": 19425
    },
    {
      "entropy": 5.178639507293701,
      "epoch": 1.6229535583027062,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004737708766826525,
      "loss": 4.8533,
      "mean_token_accuracy": 0.21071853041648864,
      "num_tokens": 35746009.0,
      "step": 19430
    },
    {
      "entropy": 5.156844997406006,
      "epoch": 1.6233711994654194,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004737569256102142,
      "loss": 4.7808,
      "mean_token_accuracy": 0.21557967960834504,
      "num_tokens": 35754012.0,
      "step": 19435
    },
    {
      "entropy": 5.069889736175537,
      "epoch": 1.6237888406281322,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047374297105829663,
      "loss": 4.9577,
      "mean_token_accuracy": 0.2127257242798805,
      "num_tokens": 35763709.0,
      "step": 19440
    },
    {
      "entropy": 5.135011816024781,
      "epoch": 1.6242064817908453,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047372901302714413,
      "loss": 4.9842,
      "mean_token_accuracy": 0.20786202400922776,
      "num_tokens": 35773457.0,
      "step": 19445
    },
    {
      "entropy": 5.076495790481568,
      "epoch": 1.6246241229535583,
      "grad_norm": 1.25,
      "learning_rate": 0.00047371505151700105,
      "loss": 4.8244,
      "mean_token_accuracy": 0.2159836247563362,
      "num_tokens": 35781713.0,
      "step": 19450
    },
    {
      "entropy": 5.113220882415772,
      "epoch": 1.6250417641162713,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047370108652811186,
      "loss": 4.8084,
      "mean_token_accuracy": 0.207745198905468,
      "num_tokens": 35790796.0,
      "step": 19455
    },
    {
      "entropy": 5.116020202636719,
      "epoch": 1.6254594052789844,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004736871180607209,
      "loss": 4.8296,
      "mean_token_accuracy": 0.2101014107465744,
      "num_tokens": 35799919.0,
      "step": 19460
    },
    {
      "entropy": 5.246385002136231,
      "epoch": 1.6258770464416972,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047367314611507276,
      "loss": 5.0168,
      "mean_token_accuracy": 0.2044241949915886,
      "num_tokens": 35809642.0,
      "step": 19465
    },
    {
      "entropy": 5.199395990371704,
      "epoch": 1.6262946876044104,
      "grad_norm": 1.125,
      "learning_rate": 0.00047365917069141215,
      "loss": 4.931,
      "mean_token_accuracy": 0.2059132069349289,
      "num_tokens": 35819020.0,
      "step": 19470
    },
    {
      "entropy": 5.252197742462158,
      "epoch": 1.6267123287671232,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004736451917899835,
      "loss": 4.9329,
      "mean_token_accuracy": 0.20503272563219072,
      "num_tokens": 35827657.0,
      "step": 19475
    },
    {
      "entropy": 5.105827474594117,
      "epoch": 1.6271299699298363,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004736312094110317,
      "loss": 4.9132,
      "mean_token_accuracy": 0.2091587707400322,
      "num_tokens": 35836727.0,
      "step": 19480
    },
    {
      "entropy": 5.068324279785156,
      "epoch": 1.6275476110925493,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004736172235548016,
      "loss": 4.872,
      "mean_token_accuracy": 0.20946940630674363,
      "num_tokens": 35846251.0,
      "step": 19485
    },
    {
      "entropy": 5.125717544555664,
      "epoch": 1.6279652522552621,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004736032342215378,
      "loss": 4.8272,
      "mean_token_accuracy": 0.2111115425825119,
      "num_tokens": 35855736.0,
      "step": 19490
    },
    {
      "entropy": 5.112504291534424,
      "epoch": 1.6283828934179754,
      "grad_norm": 1.375,
      "learning_rate": 0.00047358924141148543,
      "loss": 4.8594,
      "mean_token_accuracy": 0.22688998132944108,
      "num_tokens": 35864630.0,
      "step": 19495
    },
    {
      "entropy": 5.1698822498321535,
      "epoch": 1.6288005345806882,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047357524512488925,
      "loss": 4.9417,
      "mean_token_accuracy": 0.2055428668856621,
      "num_tokens": 35874021.0,
      "step": 19500
    },
    {
      "entropy": 5.171698760986328,
      "epoch": 1.6292181757434012,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004735612453619944,
      "loss": 4.8935,
      "mean_token_accuracy": 0.21254346817731856,
      "num_tokens": 35882696.0,
      "step": 19505
    },
    {
      "entropy": 5.255160474777222,
      "epoch": 1.6296358169061143,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000473547242123046,
      "loss": 5.0315,
      "mean_token_accuracy": 0.20309774130582808,
      "num_tokens": 35892887.0,
      "step": 19510
    },
    {
      "entropy": 5.118358469009399,
      "epoch": 1.6300534580688273,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004735332354082891,
      "loss": 4.8204,
      "mean_token_accuracy": 0.2193591445684433,
      "num_tokens": 35901547.0,
      "step": 19515
    },
    {
      "entropy": 5.206795787811279,
      "epoch": 1.6304710992315403,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047351922521796894,
      "loss": 5.0162,
      "mean_token_accuracy": 0.20539091676473617,
      "num_tokens": 35910982.0,
      "step": 19520
    },
    {
      "entropy": 5.17168173789978,
      "epoch": 1.6308887403942531,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004735052115523307,
      "loss": 4.9163,
      "mean_token_accuracy": 0.2087352305650711,
      "num_tokens": 35920548.0,
      "step": 19525
    },
    {
      "entropy": 5.30220947265625,
      "epoch": 1.6313063815569664,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004734911944116199,
      "loss": 4.9307,
      "mean_token_accuracy": 0.20653117597103118,
      "num_tokens": 35930371.0,
      "step": 19530
    },
    {
      "entropy": 5.147643661499023,
      "epoch": 1.6317240227196792,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00047347717379608184,
      "loss": 4.9736,
      "mean_token_accuracy": 0.19930318742990494,
      "num_tokens": 35939724.0,
      "step": 19535
    },
    {
      "entropy": 5.239553642272949,
      "epoch": 1.6321416638823922,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047346314970596185,
      "loss": 5.1024,
      "mean_token_accuracy": 0.19324192106723787,
      "num_tokens": 35949432.0,
      "step": 19540
    },
    {
      "entropy": 5.3119207382202145,
      "epoch": 1.6325593050451053,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047344912214150565,
      "loss": 5.0341,
      "mean_token_accuracy": 0.20220204889774324,
      "num_tokens": 35958687.0,
      "step": 19545
    },
    {
      "entropy": 5.190884780883789,
      "epoch": 1.632976946207818,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004734350911029586,
      "loss": 4.9743,
      "mean_token_accuracy": 0.20219072848558425,
      "num_tokens": 35967875.0,
      "step": 19550
    },
    {
      "entropy": 5.152971839904785,
      "epoch": 1.6333945873705313,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047342105659056645,
      "loss": 4.8634,
      "mean_token_accuracy": 0.2155694842338562,
      "num_tokens": 35977561.0,
      "step": 19555
    },
    {
      "entropy": 5.129801416397095,
      "epoch": 1.6338122285332441,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004734070186045749,
      "loss": 4.8711,
      "mean_token_accuracy": 0.21057706326246262,
      "num_tokens": 35986733.0,
      "step": 19560
    },
    {
      "entropy": 5.127779483795166,
      "epoch": 1.6342298696959572,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004733929771452296,
      "loss": 4.8606,
      "mean_token_accuracy": 0.21265952289104462,
      "num_tokens": 35995779.0,
      "step": 19565
    },
    {
      "entropy": 5.140395879745483,
      "epoch": 1.6346475108586702,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004733789322127764,
      "loss": 4.9894,
      "mean_token_accuracy": 0.20927112847566604,
      "num_tokens": 36004601.0,
      "step": 19570
    },
    {
      "entropy": 5.088112306594849,
      "epoch": 1.6350651520213833,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004733648838074613,
      "loss": 4.8698,
      "mean_token_accuracy": 0.2096273794770241,
      "num_tokens": 36014356.0,
      "step": 19575
    },
    {
      "entropy": 5.119970083236694,
      "epoch": 1.6354827931840963,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047335083192953,
      "loss": 4.8187,
      "mean_token_accuracy": 0.2169472575187683,
      "num_tokens": 36023084.0,
      "step": 19580
    },
    {
      "entropy": 5.149571228027344,
      "epoch": 1.635900434346809,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004733367765792288,
      "loss": 4.8858,
      "mean_token_accuracy": 0.2140128120779991,
      "num_tokens": 36031462.0,
      "step": 19585
    },
    {
      "entropy": 5.128756713867188,
      "epoch": 1.6363180755095224,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00047332271775680346,
      "loss": 4.8406,
      "mean_token_accuracy": 0.21316713243722915,
      "num_tokens": 36041107.0,
      "step": 19590
    },
    {
      "entropy": 5.157708168029785,
      "epoch": 1.6367357166722352,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004733086554625004,
      "loss": 4.9253,
      "mean_token_accuracy": 0.20980695337057115,
      "num_tokens": 36050026.0,
      "step": 19595
    },
    {
      "entropy": 5.133834028244019,
      "epoch": 1.6371533578349482,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004732945896965654,
      "loss": 4.9298,
      "mean_token_accuracy": 0.20855020880699157,
      "num_tokens": 36058858.0,
      "step": 19600
    },
    {
      "entropy": 5.193630981445312,
      "epoch": 1.6375709989976612,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047328052045924506,
      "loss": 4.9914,
      "mean_token_accuracy": 0.19830229133367538,
      "num_tokens": 36067995.0,
      "step": 19605
    },
    {
      "entropy": 5.141588258743286,
      "epoch": 1.637988640160374,
      "grad_norm": 1.25,
      "learning_rate": 0.0004732664477507855,
      "loss": 4.8808,
      "mean_token_accuracy": 0.20923528969287872,
      "num_tokens": 36077425.0,
      "step": 19610
    },
    {
      "entropy": 5.149539661407471,
      "epoch": 1.6384062813230873,
      "grad_norm": 1.25,
      "learning_rate": 0.00047325237157143314,
      "loss": 4.8843,
      "mean_token_accuracy": 0.20192110389471055,
      "num_tokens": 36086667.0,
      "step": 19615
    },
    {
      "entropy": 5.158013916015625,
      "epoch": 1.6388239224858,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00047323829192143436,
      "loss": 4.8544,
      "mean_token_accuracy": 0.2186214342713356,
      "num_tokens": 36095199.0,
      "step": 19620
    },
    {
      "entropy": 5.14721040725708,
      "epoch": 1.6392415636485131,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047322420880103566,
      "loss": 4.8494,
      "mean_token_accuracy": 0.20959951132535934,
      "num_tokens": 36104587.0,
      "step": 19625
    },
    {
      "entropy": 5.200499534606934,
      "epoch": 1.6396592048112262,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004732101222104836,
      "loss": 4.8867,
      "mean_token_accuracy": 0.21146339774131775,
      "num_tokens": 36114017.0,
      "step": 19630
    },
    {
      "entropy": 5.1389607906341555,
      "epoch": 1.6400768459739392,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004731960321500247,
      "loss": 4.9234,
      "mean_token_accuracy": 0.2064346641302109,
      "num_tokens": 36122301.0,
      "step": 19635
    },
    {
      "entropy": 5.079808521270752,
      "epoch": 1.6404944871366522,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004731819386199058,
      "loss": 4.8888,
      "mean_token_accuracy": 0.2150021091103554,
      "num_tokens": 36131538.0,
      "step": 19640
    },
    {
      "entropy": 5.183872890472412,
      "epoch": 1.640912128299365,
      "grad_norm": 1.25,
      "learning_rate": 0.0004731678416203734,
      "loss": 4.8868,
      "mean_token_accuracy": 0.2025318443775177,
      "num_tokens": 36140472.0,
      "step": 19645
    },
    {
      "entropy": 5.164467859268188,
      "epoch": 1.6413297694620783,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047315374115167446,
      "loss": 4.8728,
      "mean_token_accuracy": 0.2170603781938553,
      "num_tokens": 36148947.0,
      "step": 19650
    },
    {
      "entropy": 5.094189262390136,
      "epoch": 1.6417474106247911,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004731396372140558,
      "loss": 4.8342,
      "mean_token_accuracy": 0.21215320378541946,
      "num_tokens": 36157895.0,
      "step": 19655
    },
    {
      "entropy": 5.122620344161987,
      "epoch": 1.6421650517875042,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004731255298077642,
      "loss": 4.8986,
      "mean_token_accuracy": 0.21874501407146454,
      "num_tokens": 36168170.0,
      "step": 19660
    },
    {
      "entropy": 5.165857791900635,
      "epoch": 1.6425826929502172,
      "grad_norm": 1.40625,
      "learning_rate": 0.00047311141893304675,
      "loss": 4.9692,
      "mean_token_accuracy": 0.20249757021665574,
      "num_tokens": 36176864.0,
      "step": 19665
    },
    {
      "entropy": 5.271299695968628,
      "epoch": 1.64300033411293,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004730973045901505,
      "loss": 4.9495,
      "mean_token_accuracy": 0.21566179394721985,
      "num_tokens": 36186334.0,
      "step": 19670
    },
    {
      "entropy": 5.193755292892456,
      "epoch": 1.6434179752756433,
      "grad_norm": 1.3125,
      "learning_rate": 0.00047308318677932243,
      "loss": 4.9504,
      "mean_token_accuracy": 0.21028893291950226,
      "num_tokens": 36194688.0,
      "step": 19675
    },
    {
      "entropy": 5.065887689590454,
      "epoch": 1.643835616438356,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004730690655008097,
      "loss": 4.9038,
      "mean_token_accuracy": 0.20543716549873353,
      "num_tokens": 36203962.0,
      "step": 19680
    },
    {
      "entropy": 5.19875807762146,
      "epoch": 1.644253257601069,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004730549407548597,
      "loss": 4.96,
      "mean_token_accuracy": 0.19928767383098603,
      "num_tokens": 36212780.0,
      "step": 19685
    },
    {
      "entropy": 5.222076845169068,
      "epoch": 1.6446708987637821,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047304081254171944,
      "loss": 4.9391,
      "mean_token_accuracy": 0.20748526304960252,
      "num_tokens": 36221744.0,
      "step": 19690
    },
    {
      "entropy": 5.157289171218872,
      "epoch": 1.6450885399264952,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004730266808616364,
      "loss": 5.0343,
      "mean_token_accuracy": 0.20963488668203353,
      "num_tokens": 36231261.0,
      "step": 19695
    },
    {
      "entropy": 5.142521858215332,
      "epoch": 1.6455061810892082,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000473012545714858,
      "loss": 4.9318,
      "mean_token_accuracy": 0.21175138652324677,
      "num_tokens": 36240599.0,
      "step": 19700
    },
    {
      "entropy": 5.055597877502441,
      "epoch": 1.645923822251921,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047299840710163157,
      "loss": 4.7865,
      "mean_token_accuracy": 0.21832754462957382,
      "num_tokens": 36249402.0,
      "step": 19705
    },
    {
      "entropy": 5.141700220108032,
      "epoch": 1.6463414634146343,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004729842650222047,
      "loss": 4.8378,
      "mean_token_accuracy": 0.2179916352033615,
      "num_tokens": 36258379.0,
      "step": 19710
    },
    {
      "entropy": 5.1119551181793215,
      "epoch": 1.646759104577347,
      "grad_norm": 1.25,
      "learning_rate": 0.00047297011947682494,
      "loss": 4.9205,
      "mean_token_accuracy": 0.21128058135509492,
      "num_tokens": 36267393.0,
      "step": 19715
    },
    {
      "entropy": 5.220143556594849,
      "epoch": 1.6471767457400601,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047295597046574,
      "loss": 4.9511,
      "mean_token_accuracy": 0.20820419639348983,
      "num_tokens": 36277542.0,
      "step": 19720
    },
    {
      "entropy": 5.172076082229614,
      "epoch": 1.6475943869027732,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004729418179891975,
      "loss": 4.9411,
      "mean_token_accuracy": 0.2035657599568367,
      "num_tokens": 36287623.0,
      "step": 19725
    },
    {
      "entropy": 5.091217756271362,
      "epoch": 1.6480120280654862,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004729276620474452,
      "loss": 4.848,
      "mean_token_accuracy": 0.2145408272743225,
      "num_tokens": 36295828.0,
      "step": 19730
    },
    {
      "entropy": 5.1092675685882565,
      "epoch": 1.6484296692281992,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047291350264073095,
      "loss": 4.9109,
      "mean_token_accuracy": 0.2054639935493469,
      "num_tokens": 36304825.0,
      "step": 19735
    },
    {
      "entropy": 5.228623342514038,
      "epoch": 1.648847310390912,
      "grad_norm": 1.28125,
      "learning_rate": 0.00047289933976930263,
      "loss": 4.9559,
      "mean_token_accuracy": 0.2053511843085289,
      "num_tokens": 36313691.0,
      "step": 19740
    },
    {
      "entropy": 5.180575323104859,
      "epoch": 1.6492649515536253,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004728851734334081,
      "loss": 4.8905,
      "mean_token_accuracy": 0.20886103957891464,
      "num_tokens": 36323291.0,
      "step": 19745
    },
    {
      "entropy": 5.102079248428344,
      "epoch": 1.649682592716338,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004728710036332954,
      "loss": 4.8974,
      "mean_token_accuracy": 0.2091674491763115,
      "num_tokens": 36332859.0,
      "step": 19750
    },
    {
      "entropy": 5.1291382789611815,
      "epoch": 1.6501002338790511,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047285683036921275,
      "loss": 4.9296,
      "mean_token_accuracy": 0.21231217682361603,
      "num_tokens": 36342049.0,
      "step": 19755
    },
    {
      "entropy": 5.095316171646118,
      "epoch": 1.6505178750417642,
      "grad_norm": 1.265625,
      "learning_rate": 0.00047284265364140794,
      "loss": 4.8678,
      "mean_token_accuracy": 0.20943665206432344,
      "num_tokens": 36351512.0,
      "step": 19760
    },
    {
      "entropy": 5.178553771972656,
      "epoch": 1.650935516204477,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004728284734501295,
      "loss": 4.9716,
      "mean_token_accuracy": 0.20632892996072769,
      "num_tokens": 36361923.0,
      "step": 19765
    },
    {
      "entropy": 5.172000551223755,
      "epoch": 1.6513531573671902,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004728142897956253,
      "loss": 4.9604,
      "mean_token_accuracy": 0.20567336827516555,
      "num_tokens": 36370543.0,
      "step": 19770
    },
    {
      "entropy": 5.079887533187867,
      "epoch": 1.651770798529903,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000472800102678144,
      "loss": 4.8527,
      "mean_token_accuracy": 0.21997522562742233,
      "num_tokens": 36378465.0,
      "step": 19775
    },
    {
      "entropy": 5.201137924194336,
      "epoch": 1.652188439692616,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047278591209793384,
      "loss": 4.9458,
      "mean_token_accuracy": 0.20291245281696318,
      "num_tokens": 36387540.0,
      "step": 19780
    },
    {
      "entropy": 5.172557163238525,
      "epoch": 1.6526060808553291,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004727717180552431,
      "loss": 4.9378,
      "mean_token_accuracy": 0.20175271928310395,
      "num_tokens": 36397859.0,
      "step": 19785
    },
    {
      "entropy": 5.11392993927002,
      "epoch": 1.6530237220180422,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047275752055032043,
      "loss": 4.8269,
      "mean_token_accuracy": 0.2185378834605217,
      "num_tokens": 36406955.0,
      "step": 19790
    },
    {
      "entropy": 5.091060304641724,
      "epoch": 1.6534413631807552,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047274331958341434,
      "loss": 4.8256,
      "mean_token_accuracy": 0.21611739546060563,
      "num_tokens": 36415499.0,
      "step": 19795
    },
    {
      "entropy": 5.090917158126831,
      "epoch": 1.653859004343468,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004727291151547733,
      "loss": 4.8963,
      "mean_token_accuracy": 0.20433201789855956,
      "num_tokens": 36424348.0,
      "step": 19800
    },
    {
      "entropy": 5.071996879577637,
      "epoch": 1.6542766455061813,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004727149072646463,
      "loss": 4.8234,
      "mean_token_accuracy": 0.2115314394235611,
      "num_tokens": 36433134.0,
      "step": 19805
    },
    {
      "entropy": 5.021313095092774,
      "epoch": 1.654694286668894,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047270069591328175,
      "loss": 4.8007,
      "mean_token_accuracy": 0.21865687221288682,
      "num_tokens": 36442300.0,
      "step": 19810
    },
    {
      "entropy": 5.197952508926392,
      "epoch": 1.655111927831607,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047268648110092857,
      "loss": 4.9857,
      "mean_token_accuracy": 0.2005005568265915,
      "num_tokens": 36452265.0,
      "step": 19815
    },
    {
      "entropy": 5.188908576965332,
      "epoch": 1.6555295689943201,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00047267226282783555,
      "loss": 4.8953,
      "mean_token_accuracy": 0.21273574233055115,
      "num_tokens": 36462009.0,
      "step": 19820
    },
    {
      "entropy": 5.225632381439209,
      "epoch": 1.655947210157033,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004726580410942516,
      "loss": 5.023,
      "mean_token_accuracy": 0.2065310597419739,
      "num_tokens": 36470799.0,
      "step": 19825
    },
    {
      "entropy": 5.202947664260864,
      "epoch": 1.6563648513197462,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047264381590042575,
      "loss": 4.9114,
      "mean_token_accuracy": 0.20400175005197524,
      "num_tokens": 36479362.0,
      "step": 19830
    },
    {
      "entropy": 5.119249725341797,
      "epoch": 1.656782492482459,
      "grad_norm": 1.125,
      "learning_rate": 0.000472629587246607,
      "loss": 4.8336,
      "mean_token_accuracy": 0.20862033665180207,
      "num_tokens": 36487773.0,
      "step": 19835
    },
    {
      "entropy": 5.014200973510742,
      "epoch": 1.657200133645172,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047261535513304446,
      "loss": 4.8016,
      "mean_token_accuracy": 0.21748800426721573,
      "num_tokens": 36496133.0,
      "step": 19840
    },
    {
      "entropy": 5.149106121063232,
      "epoch": 1.657617774807885,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047260111955998725,
      "loss": 4.9405,
      "mean_token_accuracy": 0.20824000537395476,
      "num_tokens": 36505392.0,
      "step": 19845
    },
    {
      "entropy": 5.206650018692017,
      "epoch": 1.6580354159705981,
      "grad_norm": 1.453125,
      "learning_rate": 0.00047258688052768456,
      "loss": 4.9343,
      "mean_token_accuracy": 0.20551020354032518,
      "num_tokens": 36514876.0,
      "step": 19850
    },
    {
      "entropy": 5.183324146270752,
      "epoch": 1.6584530571333111,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004725726380363857,
      "loss": 4.867,
      "mean_token_accuracy": 0.2123461589217186,
      "num_tokens": 36523828.0,
      "step": 19855
    },
    {
      "entropy": 5.136002159118652,
      "epoch": 1.658870698296024,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00047255839208634005,
      "loss": 4.9252,
      "mean_token_accuracy": 0.20899859815835953,
      "num_tokens": 36532263.0,
      "step": 19860
    },
    {
      "entropy": 5.10527458190918,
      "epoch": 1.6592883394587372,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047254414267779686,
      "loss": 4.8262,
      "mean_token_accuracy": 0.21146710813045502,
      "num_tokens": 36541015.0,
      "step": 19865
    },
    {
      "entropy": 5.1017824649810795,
      "epoch": 1.65970598062145,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047252988981100576,
      "loss": 4.9146,
      "mean_token_accuracy": 0.20390667617321015,
      "num_tokens": 36550254.0,
      "step": 19870
    },
    {
      "entropy": 5.176391124725342,
      "epoch": 1.660123621784163,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004725156334862161,
      "loss": 4.9656,
      "mean_token_accuracy": 0.1997327908873558,
      "num_tokens": 36560941.0,
      "step": 19875
    },
    {
      "entropy": 5.20842719078064,
      "epoch": 1.660541262946876,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047250137370367755,
      "loss": 4.9665,
      "mean_token_accuracy": 0.20063198059797288,
      "num_tokens": 36570443.0,
      "step": 19880
    },
    {
      "entropy": 5.19212908744812,
      "epoch": 1.660958904109589,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047248711046363976,
      "loss": 4.9321,
      "mean_token_accuracy": 0.2074078842997551,
      "num_tokens": 36580532.0,
      "step": 19885
    },
    {
      "entropy": 5.169151782989502,
      "epoch": 1.6613765452723022,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004724728437663523,
      "loss": 4.8891,
      "mean_token_accuracy": 0.21165401339530945,
      "num_tokens": 36589359.0,
      "step": 19890
    },
    {
      "entropy": 5.176833772659302,
      "epoch": 1.661794186435015,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047245857361206506,
      "loss": 4.9221,
      "mean_token_accuracy": 0.20540316104888917,
      "num_tokens": 36598229.0,
      "step": 19895
    },
    {
      "entropy": 5.128072929382324,
      "epoch": 1.662211827597728,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047244430000102785,
      "loss": 4.8675,
      "mean_token_accuracy": 0.21026061922311784,
      "num_tokens": 36607655.0,
      "step": 19900
    },
    {
      "entropy": 5.232269382476806,
      "epoch": 1.662629468760441,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004724300229334905,
      "loss": 4.947,
      "mean_token_accuracy": 0.19815520644187928,
      "num_tokens": 36618060.0,
      "step": 19905
    },
    {
      "entropy": 5.150904989242553,
      "epoch": 1.663047109923154,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047241574240970284,
      "loss": 4.8667,
      "mean_token_accuracy": 0.21048215180635452,
      "num_tokens": 36627825.0,
      "step": 19910
    },
    {
      "entropy": 5.078194570541382,
      "epoch": 1.663464751085867,
      "grad_norm": 1.265625,
      "learning_rate": 0.00047240145842991513,
      "loss": 4.8588,
      "mean_token_accuracy": 0.21488053500652313,
      "num_tokens": 36635971.0,
      "step": 19915
    },
    {
      "entropy": 5.044627141952515,
      "epoch": 1.66388239224858,
      "grad_norm": 1.265625,
      "learning_rate": 0.00047238717099437713,
      "loss": 4.7451,
      "mean_token_accuracy": 0.21423327177762985,
      "num_tokens": 36644340.0,
      "step": 19920
    },
    {
      "entropy": 5.16188154220581,
      "epoch": 1.6643000334112932,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004723728801033392,
      "loss": 4.8997,
      "mean_token_accuracy": 0.20244514644145967,
      "num_tokens": 36654371.0,
      "step": 19925
    },
    {
      "entropy": 5.088039302825928,
      "epoch": 1.664717674574006,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004723585857570514,
      "loss": 4.7753,
      "mean_token_accuracy": 0.22335205972194672,
      "num_tokens": 36663633.0,
      "step": 19930
    },
    {
      "entropy": 5.061056852340698,
      "epoch": 1.665135315736719,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000472344287955764,
      "loss": 4.7808,
      "mean_token_accuracy": 0.22225152850151061,
      "num_tokens": 36673004.0,
      "step": 19935
    },
    {
      "entropy": 5.017529726028442,
      "epoch": 1.665552956899432,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004723299866997273,
      "loss": 4.8073,
      "mean_token_accuracy": 0.22442986071109772,
      "num_tokens": 36681323.0,
      "step": 19940
    },
    {
      "entropy": 5.201443910598755,
      "epoch": 1.6659705980621449,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004723156819891917,
      "loss": 4.9941,
      "mean_token_accuracy": 0.19492224901914595,
      "num_tokens": 36691128.0,
      "step": 19945
    },
    {
      "entropy": 5.153374814987183,
      "epoch": 1.6663882392248581,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047230137382440747,
      "loss": 4.8681,
      "mean_token_accuracy": 0.21290563493967057,
      "num_tokens": 36699759.0,
      "step": 19950
    },
    {
      "entropy": 5.050519847869873,
      "epoch": 1.666805880387571,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004722870622056253,
      "loss": 4.9228,
      "mean_token_accuracy": 0.20601690411567689,
      "num_tokens": 36708723.0,
      "step": 19955
    },
    {
      "entropy": 5.138240909576416,
      "epoch": 1.667223521550284,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004722727471330956,
      "loss": 4.8381,
      "mean_token_accuracy": 0.20698059797286988,
      "num_tokens": 36717488.0,
      "step": 19960
    },
    {
      "entropy": 5.034453535079956,
      "epoch": 1.667641162712997,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000472258428607069,
      "loss": 4.8256,
      "mean_token_accuracy": 0.21875387728214263,
      "num_tokens": 36726477.0,
      "step": 19965
    },
    {
      "entropy": 5.109641361236572,
      "epoch": 1.66805880387571,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004722441066277961,
      "loss": 4.7972,
      "mean_token_accuracy": 0.21352407038211824,
      "num_tokens": 36734676.0,
      "step": 19970
    },
    {
      "entropy": 5.151736927032471,
      "epoch": 1.668476445038423,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047222978119552776,
      "loss": 4.9564,
      "mean_token_accuracy": 0.20567291527986525,
      "num_tokens": 36744425.0,
      "step": 19975
    },
    {
      "entropy": 5.15897126197815,
      "epoch": 1.6688940862011359,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047221545231051473,
      "loss": 4.9031,
      "mean_token_accuracy": 0.2061668664216995,
      "num_tokens": 36754113.0,
      "step": 19980
    },
    {
      "entropy": 5.114077091217041,
      "epoch": 1.6693117273638491,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004722011199730078,
      "loss": 4.8435,
      "mean_token_accuracy": 0.21394485086202622,
      "num_tokens": 36763013.0,
      "step": 19985
    },
    {
      "entropy": 5.098403835296631,
      "epoch": 1.669729368526562,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004721867841832579,
      "loss": 4.913,
      "mean_token_accuracy": 0.20993608981370926,
      "num_tokens": 36772115.0,
      "step": 19990
    },
    {
      "entropy": 5.182243490219117,
      "epoch": 1.670147009689275,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004721724449415159,
      "loss": 4.9277,
      "mean_token_accuracy": 0.20289765149354935,
      "num_tokens": 36781210.0,
      "step": 19995
    },
    {
      "entropy": 5.250446605682373,
      "epoch": 1.670564650851988,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047215810224803304,
      "loss": 4.9911,
      "mean_token_accuracy": 0.20302424281835557,
      "num_tokens": 36791131.0,
      "step": 20000
    },
    {
      "entropy": 5.089558076858521,
      "epoch": 1.6709822920147008,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047214375610306016,
      "loss": 4.7663,
      "mean_token_accuracy": 0.2177559867501259,
      "num_tokens": 36800618.0,
      "step": 20005
    },
    {
      "entropy": 5.125179243087769,
      "epoch": 1.671399933177414,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047212940650684853,
      "loss": 4.9606,
      "mean_token_accuracy": 0.2021126016974449,
      "num_tokens": 36809995.0,
      "step": 20010
    },
    {
      "entropy": 5.195771551132202,
      "epoch": 1.671817574340127,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004721150534596495,
      "loss": 4.9327,
      "mean_token_accuracy": 0.20279413014650344,
      "num_tokens": 36818492.0,
      "step": 20015
    },
    {
      "entropy": 5.168107557296753,
      "epoch": 1.67223521550284,
      "grad_norm": 1.15625,
      "learning_rate": 0.000472100696961714,
      "loss": 4.887,
      "mean_token_accuracy": 0.21282435804605485,
      "num_tokens": 36827280.0,
      "step": 20020
    },
    {
      "entropy": 5.149571418762207,
      "epoch": 1.672652856665553,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004720863370132937,
      "loss": 4.9568,
      "mean_token_accuracy": 0.20722600668668748,
      "num_tokens": 36837931.0,
      "step": 20025
    },
    {
      "entropy": 5.180664348602295,
      "epoch": 1.673070497828266,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004720719736146398,
      "loss": 4.9799,
      "mean_token_accuracy": 0.20345384180545806,
      "num_tokens": 36847581.0,
      "step": 20030
    },
    {
      "entropy": 5.094317960739136,
      "epoch": 1.673488138990979,
      "grad_norm": 1.265625,
      "learning_rate": 0.00047205760676600377,
      "loss": 4.8914,
      "mean_token_accuracy": 0.20840903371572495,
      "num_tokens": 36857147.0,
      "step": 20035
    },
    {
      "entropy": 5.145553922653198,
      "epoch": 1.6739057801536918,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004720432364676372,
      "loss": 4.874,
      "mean_token_accuracy": 0.2097795471549034,
      "num_tokens": 36866105.0,
      "step": 20040
    },
    {
      "entropy": 5.174008893966675,
      "epoch": 1.674323421316405,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004720288627197915,
      "loss": 4.8447,
      "mean_token_accuracy": 0.22163391709327698,
      "num_tokens": 36874631.0,
      "step": 20045
    },
    {
      "entropy": 5.093870019912719,
      "epoch": 1.674741062479118,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004720144855227185,
      "loss": 4.832,
      "mean_token_accuracy": 0.2214858815073967,
      "num_tokens": 36884264.0,
      "step": 20050
    },
    {
      "entropy": 5.201965761184693,
      "epoch": 1.675158703641831,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004720001048766696,
      "loss": 4.9028,
      "mean_token_accuracy": 0.20594544112682342,
      "num_tokens": 36893143.0,
      "step": 20055
    },
    {
      "entropy": 5.221694993972778,
      "epoch": 1.675576344804544,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004719857207818969,
      "loss": 4.9823,
      "mean_token_accuracy": 0.20468370616436005,
      "num_tokens": 36903026.0,
      "step": 20060
    },
    {
      "entropy": 5.193270587921143,
      "epoch": 1.6759939859672568,
      "grad_norm": 1.25,
      "learning_rate": 0.000471971333238652,
      "loss": 4.9977,
      "mean_token_accuracy": 0.2040984272956848,
      "num_tokens": 36912737.0,
      "step": 20065
    },
    {
      "entropy": 5.202407503128052,
      "epoch": 1.67641162712997,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047195694224718684,
      "loss": 4.9609,
      "mean_token_accuracy": 0.20592615157365798,
      "num_tokens": 36921148.0,
      "step": 20070
    },
    {
      "entropy": 5.063308238983154,
      "epoch": 1.6768292682926829,
      "grad_norm": 1.125,
      "learning_rate": 0.00047194254780775325,
      "loss": 4.7873,
      "mean_token_accuracy": 0.22024336010217666,
      "num_tokens": 36930043.0,
      "step": 20075
    },
    {
      "entropy": 5.170555543899536,
      "epoch": 1.677246909455396,
      "grad_norm": 1.25,
      "learning_rate": 0.0004719281499206034,
      "loss": 4.9688,
      "mean_token_accuracy": 0.2022878810763359,
      "num_tokens": 36939124.0,
      "step": 20080
    },
    {
      "entropy": 5.231988096237183,
      "epoch": 1.677664550618109,
      "grad_norm": 1.28125,
      "learning_rate": 0.00047191374858598917,
      "loss": 5.0165,
      "mean_token_accuracy": 0.20591458678245544,
      "num_tokens": 36948392.0,
      "step": 20085
    },
    {
      "entropy": 5.1769157409667965,
      "epoch": 1.678082191780822,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004718993438041628,
      "loss": 4.9053,
      "mean_token_accuracy": 0.2088561475276947,
      "num_tokens": 36958019.0,
      "step": 20090
    },
    {
      "entropy": 5.130549192428589,
      "epoch": 1.678499832943535,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047188493557537634,
      "loss": 4.8582,
      "mean_token_accuracy": 0.2067590683698654,
      "num_tokens": 36966904.0,
      "step": 20095
    },
    {
      "entropy": 5.210665273666382,
      "epoch": 1.6789174741062478,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004718705238998822,
      "loss": 4.9316,
      "mean_token_accuracy": 0.20695365816354752,
      "num_tokens": 36976085.0,
      "step": 20100
    },
    {
      "entropy": 5.11279706954956,
      "epoch": 1.679335115268961,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047185610877793245,
      "loss": 4.8432,
      "mean_token_accuracy": 0.21954097896814345,
      "num_tokens": 36984345.0,
      "step": 20105
    },
    {
      "entropy": 5.076624155044556,
      "epoch": 1.6797527564316739,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047184169020977965,
      "loss": 4.8679,
      "mean_token_accuracy": 0.21231902539730071,
      "num_tokens": 36993133.0,
      "step": 20110
    },
    {
      "entropy": 5.132299184799194,
      "epoch": 1.680170397594387,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047182726819567603,
      "loss": 4.9603,
      "mean_token_accuracy": 0.20526651442050933,
      "num_tokens": 37002797.0,
      "step": 20115
    },
    {
      "entropy": 5.2365028858184814,
      "epoch": 1.6805880387571,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047181284273587414,
      "loss": 4.8562,
      "mean_token_accuracy": 0.2088881716132164,
      "num_tokens": 37012598.0,
      "step": 20120
    },
    {
      "entropy": 5.1450536251068115,
      "epoch": 1.6810056799198128,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047179841383062667,
      "loss": 4.8808,
      "mean_token_accuracy": 0.21596155762672425,
      "num_tokens": 37022006.0,
      "step": 20125
    },
    {
      "entropy": 5.01507248878479,
      "epoch": 1.681423321082526,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047178398148018606,
      "loss": 4.8155,
      "mean_token_accuracy": 0.21403166502714158,
      "num_tokens": 37030713.0,
      "step": 20130
    },
    {
      "entropy": 5.081826162338257,
      "epoch": 1.6818409622452388,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004717695456848049,
      "loss": 4.8762,
      "mean_token_accuracy": 0.21597574800252914,
      "num_tokens": 37039347.0,
      "step": 20135
    },
    {
      "entropy": 5.245157957077026,
      "epoch": 1.6822586034079519,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004717551064447361,
      "loss": 4.9506,
      "mean_token_accuracy": 0.2048339366912842,
      "num_tokens": 37048010.0,
      "step": 20140
    },
    {
      "entropy": 5.090592861175537,
      "epoch": 1.682676244570665,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004717406637602322,
      "loss": 4.813,
      "mean_token_accuracy": 0.21493736356496812,
      "num_tokens": 37056517.0,
      "step": 20145
    },
    {
      "entropy": 5.077041339874268,
      "epoch": 1.683093885733378,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004717262176315462,
      "loss": 4.8627,
      "mean_token_accuracy": 0.20590561181306838,
      "num_tokens": 37065085.0,
      "step": 20150
    },
    {
      "entropy": 5.130452489852905,
      "epoch": 1.683511526896091,
      "grad_norm": 1.234375,
      "learning_rate": 0.000471711768058931,
      "loss": 4.831,
      "mean_token_accuracy": 0.21656909137964248,
      "num_tokens": 37074483.0,
      "step": 20155
    },
    {
      "entropy": 5.189768886566162,
      "epoch": 1.6839291680588038,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004716973150426395,
      "loss": 4.9289,
      "mean_token_accuracy": 0.2056008130311966,
      "num_tokens": 37083679.0,
      "step": 20160
    },
    {
      "entropy": 5.113491344451904,
      "epoch": 1.684346809221517,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047168285858292477,
      "loss": 4.8865,
      "mean_token_accuracy": 0.21124650537967682,
      "num_tokens": 37092258.0,
      "step": 20165
    },
    {
      "entropy": 5.154540967941284,
      "epoch": 1.6847644503842298,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047166839868003984,
      "loss": 4.9918,
      "mean_token_accuracy": 0.1973344385623932,
      "num_tokens": 37101100.0,
      "step": 20170
    },
    {
      "entropy": 5.240135431289673,
      "epoch": 1.6851820915469429,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047165393533423784,
      "loss": 5.0087,
      "mean_token_accuracy": 0.20308649092912673,
      "num_tokens": 37110876.0,
      "step": 20175
    },
    {
      "entropy": 5.22374210357666,
      "epoch": 1.685599732709656,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047163946854577194,
      "loss": 4.8732,
      "mean_token_accuracy": 0.21036168485879897,
      "num_tokens": 37119770.0,
      "step": 20180
    },
    {
      "entropy": 5.077211475372314,
      "epoch": 1.6860173738723687,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004716249983148955,
      "loss": 4.8672,
      "mean_token_accuracy": 0.21211252063512803,
      "num_tokens": 37129173.0,
      "step": 20185
    },
    {
      "entropy": 5.078359222412109,
      "epoch": 1.686435015035082,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004716105246418618,
      "loss": 4.8195,
      "mean_token_accuracy": 0.21616876274347305,
      "num_tokens": 37138229.0,
      "step": 20190
    },
    {
      "entropy": 5.170198678970337,
      "epoch": 1.6868526561977948,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004715960475269242,
      "loss": 4.9671,
      "mean_token_accuracy": 0.2057353898882866,
      "num_tokens": 37147792.0,
      "step": 20195
    },
    {
      "entropy": 5.189734220504761,
      "epoch": 1.6872702973605078,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004715815669703362,
      "loss": 4.8709,
      "mean_token_accuracy": 0.21579850614070892,
      "num_tokens": 37156729.0,
      "step": 20200
    },
    {
      "entropy": 5.099260520935059,
      "epoch": 1.6876879385232209,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047156708297235115,
      "loss": 4.8876,
      "mean_token_accuracy": 0.21037989109754562,
      "num_tokens": 37165678.0,
      "step": 20205
    },
    {
      "entropy": 5.233407688140869,
      "epoch": 1.6881055796859339,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004715525955332227,
      "loss": 4.9985,
      "mean_token_accuracy": 0.20750170797109604,
      "num_tokens": 37174704.0,
      "step": 20210
    },
    {
      "entropy": 5.184150695800781,
      "epoch": 1.688523220848647,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004715381046532046,
      "loss": 4.9263,
      "mean_token_accuracy": 0.2069593071937561,
      "num_tokens": 37184548.0,
      "step": 20215
    },
    {
      "entropy": 5.117223405838013,
      "epoch": 1.6889408620113597,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004715236103325503,
      "loss": 4.8498,
      "mean_token_accuracy": 0.21151609867811202,
      "num_tokens": 37193668.0,
      "step": 20220
    },
    {
      "entropy": 5.124175310134888,
      "epoch": 1.689358503174073,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047150911257151364,
      "loss": 4.8893,
      "mean_token_accuracy": 0.20818138420581817,
      "num_tokens": 37202504.0,
      "step": 20225
    },
    {
      "entropy": 5.121514701843262,
      "epoch": 1.6897761443367858,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004714946113703484,
      "loss": 4.8325,
      "mean_token_accuracy": 0.21846916973590852,
      "num_tokens": 37212140.0,
      "step": 20230
    },
    {
      "entropy": 5.189235639572144,
      "epoch": 1.6901937854994988,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004714801067293086,
      "loss": 5.024,
      "mean_token_accuracy": 0.20598167181015015,
      "num_tokens": 37221824.0,
      "step": 20235
    },
    {
      "entropy": 5.227892923355102,
      "epoch": 1.6906114266622119,
      "grad_norm": 1.125,
      "learning_rate": 0.00047146559864864787,
      "loss": 4.9787,
      "mean_token_accuracy": 0.20460404455661774,
      "num_tokens": 37231877.0,
      "step": 20240
    },
    {
      "entropy": 5.22090334892273,
      "epoch": 1.6910290678249247,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047145108712862046,
      "loss": 4.9581,
      "mean_token_accuracy": 0.20841053426265715,
      "num_tokens": 37240601.0,
      "step": 20245
    },
    {
      "entropy": 5.062189960479737,
      "epoch": 1.691446708987638,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004714365721694803,
      "loss": 4.8592,
      "mean_token_accuracy": 0.2081640839576721,
      "num_tokens": 37249645.0,
      "step": 20250
    },
    {
      "entropy": 5.100494146347046,
      "epoch": 1.6918643501503507,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047142205377148145,
      "loss": 4.8665,
      "mean_token_accuracy": 0.21324455440044404,
      "num_tokens": 37258170.0,
      "step": 20255
    },
    {
      "entropy": 5.119489240646362,
      "epoch": 1.6922819913130638,
      "grad_norm": 1.25,
      "learning_rate": 0.0004714075319348781,
      "loss": 4.9429,
      "mean_token_accuracy": 0.20820362418889998,
      "num_tokens": 37268045.0,
      "step": 20260
    },
    {
      "entropy": 5.160318899154663,
      "epoch": 1.6926996324757768,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004713930066599246,
      "loss": 4.9043,
      "mean_token_accuracy": 0.21068683713674546,
      "num_tokens": 37278558.0,
      "step": 20265
    },
    {
      "entropy": 5.141438627243042,
      "epoch": 1.6931172736384899,
      "grad_norm": 1.171875,
      "learning_rate": 0.000471378477946875,
      "loss": 4.7955,
      "mean_token_accuracy": 0.21853651553392411,
      "num_tokens": 37287119.0,
      "step": 20270
    },
    {
      "entropy": 5.128001165390015,
      "epoch": 1.6935349148012029,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004713639457959838,
      "loss": 4.9041,
      "mean_token_accuracy": 0.20813864171504975,
      "num_tokens": 37296300.0,
      "step": 20275
    },
    {
      "entropy": 5.140497159957886,
      "epoch": 1.6939525559639157,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004713494102075053,
      "loss": 4.8812,
      "mean_token_accuracy": 0.21384607255458832,
      "num_tokens": 37305291.0,
      "step": 20280
    },
    {
      "entropy": 5.1103373050689695,
      "epoch": 1.694370197126629,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004713348711816942,
      "loss": 4.8538,
      "mean_token_accuracy": 0.2144835263490677,
      "num_tokens": 37314441.0,
      "step": 20285
    },
    {
      "entropy": 5.141145372390747,
      "epoch": 1.6947878382893418,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004713203287188047,
      "loss": 4.9334,
      "mean_token_accuracy": 0.2137947455048561,
      "num_tokens": 37323516.0,
      "step": 20290
    },
    {
      "entropy": 5.169126415252686,
      "epoch": 1.6952054794520548,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004713057828190916,
      "loss": 4.8872,
      "mean_token_accuracy": 0.21406176537275315,
      "num_tokens": 37332651.0,
      "step": 20295
    },
    {
      "entropy": 5.163804578781128,
      "epoch": 1.6956231206147678,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004712912334828094,
      "loss": 4.9287,
      "mean_token_accuracy": 0.20208457857370377,
      "num_tokens": 37341580.0,
      "step": 20300
    },
    {
      "entropy": 5.153610897064209,
      "epoch": 1.6960407617774806,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047127668071021293,
      "loss": 4.9851,
      "mean_token_accuracy": 0.20382054299116134,
      "num_tokens": 37351385.0,
      "step": 20305
    },
    {
      "entropy": 5.009532642364502,
      "epoch": 1.696458402940194,
      "grad_norm": 1.4609375,
      "learning_rate": 0.000471262124501557,
      "loss": 4.7298,
      "mean_token_accuracy": 0.22358565181493759,
      "num_tokens": 37360893.0,
      "step": 20310
    },
    {
      "entropy": 5.1082127571105955,
      "epoch": 1.6968760441029067,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004712475648570962,
      "loss": 4.843,
      "mean_token_accuracy": 0.2126660481095314,
      "num_tokens": 37370301.0,
      "step": 20315
    },
    {
      "entropy": 5.167234897613525,
      "epoch": 1.6972936852656197,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047123300177708563,
      "loss": 4.9336,
      "mean_token_accuracy": 0.2081856518983841,
      "num_tokens": 37380881.0,
      "step": 20320
    },
    {
      "entropy": 5.098295259475708,
      "epoch": 1.6977113264283328,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004712184352617801,
      "loss": 4.8342,
      "mean_token_accuracy": 0.21759311258792877,
      "num_tokens": 37390167.0,
      "step": 20325
    },
    {
      "entropy": 5.222169733047485,
      "epoch": 1.6981289675910458,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047120386531143476,
      "loss": 5.0693,
      "mean_token_accuracy": 0.19849362671375276,
      "num_tokens": 37399163.0,
      "step": 20330
    },
    {
      "entropy": 5.244269275665284,
      "epoch": 1.6985466087537588,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047118929192630453,
      "loss": 4.9964,
      "mean_token_accuracy": 0.2017977386713028,
      "num_tokens": 37409047.0,
      "step": 20335
    },
    {
      "entropy": 5.131319189071656,
      "epoch": 1.6989642499164717,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004711747151066446,
      "loss": 4.812,
      "mean_token_accuracy": 0.21852529644966126,
      "num_tokens": 37417936.0,
      "step": 20340
    },
    {
      "entropy": 5.00330958366394,
      "epoch": 1.699381891079185,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047116013485271014,
      "loss": 4.7317,
      "mean_token_accuracy": 0.2194441094994545,
      "num_tokens": 37426453.0,
      "step": 20345
    },
    {
      "entropy": 5.023327970504761,
      "epoch": 1.6997995322418977,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004711455511647564,
      "loss": 4.6894,
      "mean_token_accuracy": 0.23702831715345382,
      "num_tokens": 37435353.0,
      "step": 20350
    },
    {
      "entropy": 5.106814765930176,
      "epoch": 1.7002171734046108,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047113096404303875,
      "loss": 4.9371,
      "mean_token_accuracy": 0.20927560031414033,
      "num_tokens": 37444895.0,
      "step": 20355
    },
    {
      "entropy": 5.174426412582397,
      "epoch": 1.7006348145673238,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004711163734878124,
      "loss": 4.8194,
      "mean_token_accuracy": 0.2121441438794136,
      "num_tokens": 37454038.0,
      "step": 20360
    },
    {
      "entropy": 5.113776922225952,
      "epoch": 1.7010524557300366,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004711017794993329,
      "loss": 4.9023,
      "mean_token_accuracy": 0.20844349563121795,
      "num_tokens": 37463316.0,
      "step": 20365
    },
    {
      "entropy": 5.114778757095337,
      "epoch": 1.7014700968927499,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047108718207785575,
      "loss": 4.8886,
      "mean_token_accuracy": 0.20817212760448456,
      "num_tokens": 37473341.0,
      "step": 20370
    },
    {
      "entropy": 5.181875514984131,
      "epoch": 1.7018877380554627,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004710725812236364,
      "loss": 4.9224,
      "mean_token_accuracy": 0.2125932529568672,
      "num_tokens": 37482601.0,
      "step": 20375
    },
    {
      "entropy": 5.103157663345337,
      "epoch": 1.7023053792181757,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004710579769369305,
      "loss": 4.8236,
      "mean_token_accuracy": 0.21223595589399338,
      "num_tokens": 37491304.0,
      "step": 20380
    },
    {
      "entropy": 5.02479739189148,
      "epoch": 1.7027230203808887,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004710433692179937,
      "loss": 4.8131,
      "mean_token_accuracy": 0.20990963131189347,
      "num_tokens": 37500848.0,
      "step": 20385
    },
    {
      "entropy": 5.114013624191284,
      "epoch": 1.7031406615436018,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004710287580670818,
      "loss": 4.7955,
      "mean_token_accuracy": 0.22186270356178284,
      "num_tokens": 37509649.0,
      "step": 20390
    },
    {
      "entropy": 5.124630737304687,
      "epoch": 1.7035583027063148,
      "grad_norm": 1.125,
      "learning_rate": 0.00047101414348445046,
      "loss": 4.937,
      "mean_token_accuracy": 0.20465653985738755,
      "num_tokens": 37519207.0,
      "step": 20395
    },
    {
      "entropy": 5.182248163223266,
      "epoch": 1.7039759438690276,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047099952547035564,
      "loss": 4.9328,
      "mean_token_accuracy": 0.20179950445890427,
      "num_tokens": 37528819.0,
      "step": 20400
    },
    {
      "entropy": 5.150438642501831,
      "epoch": 1.7043935850317409,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004709849040250532,
      "loss": 4.8601,
      "mean_token_accuracy": 0.21081131994724273,
      "num_tokens": 37538439.0,
      "step": 20405
    },
    {
      "entropy": 5.045848655700683,
      "epoch": 1.7048112261944537,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004709702791487991,
      "loss": 4.8312,
      "mean_token_accuracy": 0.21567142605781556,
      "num_tokens": 37547758.0,
      "step": 20410
    },
    {
      "entropy": 5.099242258071899,
      "epoch": 1.7052288673571667,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004709556508418494,
      "loss": 4.879,
      "mean_token_accuracy": 0.21421565264463424,
      "num_tokens": 37556693.0,
      "step": 20415
    },
    {
      "entropy": 5.104217863082885,
      "epoch": 1.7056465085198798,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004709410191044601,
      "loss": 4.9231,
      "mean_token_accuracy": 0.2133731096982956,
      "num_tokens": 37564631.0,
      "step": 20420
    },
    {
      "entropy": 5.105108070373535,
      "epoch": 1.7060641496825926,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004709263839368875,
      "loss": 4.8021,
      "mean_token_accuracy": 0.21565776616334914,
      "num_tokens": 37573687.0,
      "step": 20425
    },
    {
      "entropy": 5.0927979946136475,
      "epoch": 1.7064817908453058,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004709117453393876,
      "loss": 4.8673,
      "mean_token_accuracy": 0.21401216834783554,
      "num_tokens": 37582652.0,
      "step": 20430
    },
    {
      "entropy": 5.050708675384522,
      "epoch": 1.7068994320080186,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047089710331221687,
      "loss": 4.7388,
      "mean_token_accuracy": 0.21802332550287246,
      "num_tokens": 37591242.0,
      "step": 20435
    },
    {
      "entropy": 5.210809659957886,
      "epoch": 1.7073170731707317,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004708824578556315,
      "loss": 5.0841,
      "mean_token_accuracy": 0.20051846504211426,
      "num_tokens": 37601527.0,
      "step": 20440
    },
    {
      "entropy": 5.22066593170166,
      "epoch": 1.7077347143334447,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047086780896988796,
      "loss": 4.9467,
      "mean_token_accuracy": 0.20138048529624938,
      "num_tokens": 37611100.0,
      "step": 20445
    },
    {
      "entropy": 5.126105117797851,
      "epoch": 1.7081523554961577,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047085315665524257,
      "loss": 4.8692,
      "mean_token_accuracy": 0.2177979901432991,
      "num_tokens": 37619386.0,
      "step": 20450
    },
    {
      "entropy": 5.159240770339966,
      "epoch": 1.7085699966588708,
      "grad_norm": 1.21875,
      "learning_rate": 0.000470838500911952,
      "loss": 4.9952,
      "mean_token_accuracy": 0.20525255501270295,
      "num_tokens": 37628695.0,
      "step": 20455
    },
    {
      "entropy": 5.183454608917236,
      "epoch": 1.7089876378215836,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004708238417402728,
      "loss": 4.9226,
      "mean_token_accuracy": 0.2030936136841774,
      "num_tokens": 37637588.0,
      "step": 20460
    },
    {
      "entropy": 5.10713677406311,
      "epoch": 1.7094052789842968,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047080917914046146,
      "loss": 4.7885,
      "mean_token_accuracy": 0.21998721957206727,
      "num_tokens": 37646114.0,
      "step": 20465
    },
    {
      "entropy": 5.207769203186035,
      "epoch": 1.7098229201470097,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047079451311277473,
      "loss": 5.022,
      "mean_token_accuracy": 0.20086781233549117,
      "num_tokens": 37654987.0,
      "step": 20470
    },
    {
      "entropy": 5.147400760650635,
      "epoch": 1.7102405613097227,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004707798436574694,
      "loss": 4.8887,
      "mean_token_accuracy": 0.2153683304786682,
      "num_tokens": 37664715.0,
      "step": 20475
    },
    {
      "entropy": 5.149871730804444,
      "epoch": 1.7106582024724357,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047076517077480223,
      "loss": 4.9782,
      "mean_token_accuracy": 0.2102722331881523,
      "num_tokens": 37675016.0,
      "step": 20480
    },
    {
      "entropy": 5.173229646682739,
      "epoch": 1.7110758436351485,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004707504944650301,
      "loss": 4.895,
      "mean_token_accuracy": 0.21164676249027253,
      "num_tokens": 37683577.0,
      "step": 20485
    },
    {
      "entropy": 5.199729537963867,
      "epoch": 1.7114934847978618,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047073581472840997,
      "loss": 4.8927,
      "mean_token_accuracy": 0.2039148047566414,
      "num_tokens": 37692681.0,
      "step": 20490
    },
    {
      "entropy": 5.026130294799804,
      "epoch": 1.7119111259605746,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004707211315651988,
      "loss": 4.8026,
      "mean_token_accuracy": 0.21846601068973542,
      "num_tokens": 37701698.0,
      "step": 20495
    },
    {
      "entropy": 5.119742345809937,
      "epoch": 1.7123287671232876,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004707064449756536,
      "loss": 4.9032,
      "mean_token_accuracy": 0.2100774899125099,
      "num_tokens": 37710450.0,
      "step": 20500
    },
    {
      "entropy": 5.15555419921875,
      "epoch": 1.7127464082860007,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047069175496003147,
      "loss": 4.9374,
      "mean_token_accuracy": 0.2071992948651314,
      "num_tokens": 37719201.0,
      "step": 20505
    },
    {
      "entropy": 5.232965040206909,
      "epoch": 1.7131640494487137,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047067706151858967,
      "loss": 4.9773,
      "mean_token_accuracy": 0.20649203509092331,
      "num_tokens": 37728671.0,
      "step": 20510
    },
    {
      "entropy": 5.126397752761841,
      "epoch": 1.7135816906114267,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004706623646515854,
      "loss": 4.8578,
      "mean_token_accuracy": 0.21340841799974442,
      "num_tokens": 37737675.0,
      "step": 20515
    },
    {
      "entropy": 5.063002443313598,
      "epoch": 1.7139993317741395,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047064766435927585,
      "loss": 4.8706,
      "mean_token_accuracy": 0.22159244269132614,
      "num_tokens": 37746299.0,
      "step": 20520
    },
    {
      "entropy": 5.049413442611694,
      "epoch": 1.7144169729368528,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047063296064191846,
      "loss": 4.7568,
      "mean_token_accuracy": 0.21625775694847107,
      "num_tokens": 37755770.0,
      "step": 20525
    },
    {
      "entropy": 5.1061797618865965,
      "epoch": 1.7148346140995656,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047061825349977067,
      "loss": 4.9029,
      "mean_token_accuracy": 0.2110523670911789,
      "num_tokens": 37765217.0,
      "step": 20530
    },
    {
      "entropy": 5.188344573974609,
      "epoch": 1.7152522552622786,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047060354293308976,
      "loss": 4.8745,
      "mean_token_accuracy": 0.2158093646168709,
      "num_tokens": 37773891.0,
      "step": 20535
    },
    {
      "entropy": 5.11781940460205,
      "epoch": 1.7156698964249917,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004705888289421335,
      "loss": 4.8849,
      "mean_token_accuracy": 0.21233015060424804,
      "num_tokens": 37783323.0,
      "step": 20540
    },
    {
      "entropy": 5.195895862579346,
      "epoch": 1.7160875375877045,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004705741115271593,
      "loss": 4.9705,
      "mean_token_accuracy": 0.2075568065047264,
      "num_tokens": 37792436.0,
      "step": 20545
    },
    {
      "entropy": 5.143573760986328,
      "epoch": 1.7165051787504177,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004705593906884249,
      "loss": 4.8737,
      "mean_token_accuracy": 0.2124375209212303,
      "num_tokens": 37802061.0,
      "step": 20550
    },
    {
      "entropy": 5.059374141693115,
      "epoch": 1.7169228199131306,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047054466642618793,
      "loss": 4.7904,
      "mean_token_accuracy": 0.21689361929893494,
      "num_tokens": 37811101.0,
      "step": 20555
    },
    {
      "entropy": 5.202487897872925,
      "epoch": 1.7173404610758436,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047052993874070623,
      "loss": 4.9449,
      "mean_token_accuracy": 0.213174407184124,
      "num_tokens": 37820783.0,
      "step": 20560
    },
    {
      "entropy": 5.128157377243042,
      "epoch": 1.7177581022385566,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047051520763223754,
      "loss": 4.845,
      "mean_token_accuracy": 0.2075954407453537,
      "num_tokens": 37830075.0,
      "step": 20565
    },
    {
      "entropy": 5.0103864669799805,
      "epoch": 1.7181757434012697,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047050047310103984,
      "loss": 4.7544,
      "mean_token_accuracy": 0.21203645020723344,
      "num_tokens": 37838344.0,
      "step": 20570
    },
    {
      "entropy": 5.041352605819702,
      "epoch": 1.7185933845639827,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047048573514737107,
      "loss": 4.8628,
      "mean_token_accuracy": 0.20786558985710143,
      "num_tokens": 37847911.0,
      "step": 20575
    },
    {
      "entropy": 5.011209630966187,
      "epoch": 1.7190110257266955,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047047099377148916,
      "loss": 4.7933,
      "mean_token_accuracy": 0.21856043189764024,
      "num_tokens": 37857044.0,
      "step": 20580
    },
    {
      "entropy": 5.186618137359619,
      "epoch": 1.7194286668894088,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004704562489736522,
      "loss": 4.8774,
      "mean_token_accuracy": 0.21602001488208772,
      "num_tokens": 37865912.0,
      "step": 20585
    },
    {
      "entropy": 5.204092884063721,
      "epoch": 1.7198463080521216,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004704415007541184,
      "loss": 4.9634,
      "mean_token_accuracy": 0.20746199935674667,
      "num_tokens": 37875118.0,
      "step": 20590
    },
    {
      "entropy": 5.108314943313599,
      "epoch": 1.7202639492148346,
      "grad_norm": 1.25,
      "learning_rate": 0.0004704267491131458,
      "loss": 4.8256,
      "mean_token_accuracy": 0.21964281648397446,
      "num_tokens": 37882891.0,
      "step": 20595
    },
    {
      "entropy": 5.080936288833618,
      "epoch": 1.7206815903775476,
      "grad_norm": 1.25,
      "learning_rate": 0.00047041199405099273,
      "loss": 4.8923,
      "mean_token_accuracy": 0.21962003111839296,
      "num_tokens": 37892274.0,
      "step": 20600
    },
    {
      "entropy": 5.031831502914429,
      "epoch": 1.7210992315402605,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047039723556791744,
      "loss": 4.771,
      "mean_token_accuracy": 0.22279851883649826,
      "num_tokens": 37901061.0,
      "step": 20605
    },
    {
      "entropy": 5.057650470733643,
      "epoch": 1.7215168727029737,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004703824736641785,
      "loss": 4.7495,
      "mean_token_accuracy": 0.217088183760643,
      "num_tokens": 37910253.0,
      "step": 20610
    },
    {
      "entropy": 5.103163528442383,
      "epoch": 1.7219345138656865,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047036770834003405,
      "loss": 4.8854,
      "mean_token_accuracy": 0.209565669298172,
      "num_tokens": 37919453.0,
      "step": 20615
    },
    {
      "entropy": 5.033615303039551,
      "epoch": 1.7223521550283996,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00047035293959574273,
      "loss": 4.7607,
      "mean_token_accuracy": 0.21991225481033325,
      "num_tokens": 37928462.0,
      "step": 20620
    },
    {
      "entropy": 5.145063066482544,
      "epoch": 1.7227697961911126,
      "grad_norm": 1.140625,
      "learning_rate": 0.000470338167431563,
      "loss": 4.8964,
      "mean_token_accuracy": 0.21582703441381454,
      "num_tokens": 37938631.0,
      "step": 20625
    },
    {
      "entropy": 5.183991861343384,
      "epoch": 1.7231874373538256,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00047032339184775366,
      "loss": 4.9203,
      "mean_token_accuracy": 0.20074968934059143,
      "num_tokens": 37948490.0,
      "step": 20630
    },
    {
      "entropy": 4.998717451095581,
      "epoch": 1.7236050785165387,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004703086128445731,
      "loss": 4.7469,
      "mean_token_accuracy": 0.2296707198023796,
      "num_tokens": 37957861.0,
      "step": 20635
    },
    {
      "entropy": 5.028881978988648,
      "epoch": 1.7240227196792515,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004702938304222803,
      "loss": 4.8603,
      "mean_token_accuracy": 0.21455631405115128,
      "num_tokens": 37966587.0,
      "step": 20640
    },
    {
      "entropy": 5.111816787719727,
      "epoch": 1.7244403608419647,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047027904458113386,
      "loss": 4.9099,
      "mean_token_accuracy": 0.2115527629852295,
      "num_tokens": 37976295.0,
      "step": 20645
    },
    {
      "entropy": 5.325541305541992,
      "epoch": 1.7248580020046775,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004702642553213927,
      "loss": 5.122,
      "mean_token_accuracy": 0.19718281775712967,
      "num_tokens": 37985525.0,
      "step": 20650
    },
    {
      "entropy": 5.26118278503418,
      "epoch": 1.7252756431673906,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047024946264331564,
      "loss": 4.9501,
      "mean_token_accuracy": 0.20692724734544754,
      "num_tokens": 37995656.0,
      "step": 20655
    },
    {
      "entropy": 5.1081507205963135,
      "epoch": 1.7256932843301036,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004702346665471618,
      "loss": 4.8838,
      "mean_token_accuracy": 0.20888650119304658,
      "num_tokens": 38004790.0,
      "step": 20660
    },
    {
      "entropy": 5.1769153594970705,
      "epoch": 1.7261109254928164,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004702198670331901,
      "loss": 5.0047,
      "mean_token_accuracy": 0.200289486348629,
      "num_tokens": 38014244.0,
      "step": 20665
    },
    {
      "entropy": 5.112754154205322,
      "epoch": 1.7265285666555297,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047020506410165965,
      "loss": 4.8588,
      "mean_token_accuracy": 0.22581789791584014,
      "num_tokens": 38023756.0,
      "step": 20670
    },
    {
      "entropy": 5.060931587219239,
      "epoch": 1.7269462078182425,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047019025775282955,
      "loss": 4.8241,
      "mean_token_accuracy": 0.21616247594356536,
      "num_tokens": 38033402.0,
      "step": 20675
    },
    {
      "entropy": 5.107394647598267,
      "epoch": 1.7273638489809555,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000470175447986959,
      "loss": 4.8733,
      "mean_token_accuracy": 0.20645065307617189,
      "num_tokens": 38042580.0,
      "step": 20680
    },
    {
      "entropy": 5.156351518630982,
      "epoch": 1.7277814901436686,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004701606348043074,
      "loss": 4.966,
      "mean_token_accuracy": 0.20823943018913268,
      "num_tokens": 38051809.0,
      "step": 20685
    },
    {
      "entropy": 5.1146355152130125,
      "epoch": 1.7281991313063816,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004701458182051339,
      "loss": 4.7693,
      "mean_token_accuracy": 0.22028056681156158,
      "num_tokens": 38061053.0,
      "step": 20690
    },
    {
      "entropy": 5.046344470977783,
      "epoch": 1.7286167724690946,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047013099818969797,
      "loss": 4.7881,
      "mean_token_accuracy": 0.20562682598829268,
      "num_tokens": 38070190.0,
      "step": 20695
    },
    {
      "entropy": 5.073409366607666,
      "epoch": 1.7290344136318074,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000470116174758259,
      "loss": 4.9801,
      "mean_token_accuracy": 0.21251367628574372,
      "num_tokens": 38078049.0,
      "step": 20700
    },
    {
      "entropy": 5.192161321640015,
      "epoch": 1.7294520547945207,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047010134791107663,
      "loss": 4.9037,
      "mean_token_accuracy": 0.2094970315694809,
      "num_tokens": 38089130.0,
      "step": 20705
    },
    {
      "entropy": 5.116767358779907,
      "epoch": 1.7298696959572335,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047008651764841034,
      "loss": 4.7987,
      "mean_token_accuracy": 0.2138035386800766,
      "num_tokens": 38098484.0,
      "step": 20710
    },
    {
      "entropy": 5.098094892501831,
      "epoch": 1.7302873371199465,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047007168397051956,
      "loss": 4.8777,
      "mean_token_accuracy": 0.2138095572590828,
      "num_tokens": 38107982.0,
      "step": 20715
    },
    {
      "entropy": 5.125138425827027,
      "epoch": 1.7307049782826596,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047005684687766424,
      "loss": 4.8737,
      "mean_token_accuracy": 0.21628297865390778,
      "num_tokens": 38116816.0,
      "step": 20720
    },
    {
      "entropy": 5.094290113449096,
      "epoch": 1.7311226194453724,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004700420063701041,
      "loss": 4.8039,
      "mean_token_accuracy": 0.21824126690626144,
      "num_tokens": 38125752.0,
      "step": 20725
    },
    {
      "entropy": 5.0943817615509035,
      "epoch": 1.7315402606080856,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004700271624480988,
      "loss": 4.8907,
      "mean_token_accuracy": 0.21366541385650634,
      "num_tokens": 38135192.0,
      "step": 20730
    },
    {
      "entropy": 5.178084087371826,
      "epoch": 1.7319579017707984,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047001231511190826,
      "loss": 4.9441,
      "mean_token_accuracy": 0.20199714452028275,
      "num_tokens": 38144730.0,
      "step": 20735
    },
    {
      "entropy": 5.016309642791748,
      "epoch": 1.7323755429335115,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004699974643617924,
      "loss": 4.8089,
      "mean_token_accuracy": 0.21808530986309052,
      "num_tokens": 38153319.0,
      "step": 20740
    },
    {
      "entropy": 5.107643270492554,
      "epoch": 1.7327931840962245,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004699826101980112,
      "loss": 4.8334,
      "mean_token_accuracy": 0.21578191220760345,
      "num_tokens": 38162067.0,
      "step": 20745
    },
    {
      "entropy": 5.252505540847778,
      "epoch": 1.7332108252589375,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004699677526208248,
      "loss": 4.988,
      "mean_token_accuracy": 0.2053759753704071,
      "num_tokens": 38170901.0,
      "step": 20750
    },
    {
      "entropy": 5.056118822097778,
      "epoch": 1.7336284664216506,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00046995289163049317,
      "loss": 4.8912,
      "mean_token_accuracy": 0.2115262418985367,
      "num_tokens": 38179815.0,
      "step": 20755
    },
    {
      "entropy": 5.114478206634521,
      "epoch": 1.7340461075843634,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046993802722727646,
      "loss": 4.9045,
      "mean_token_accuracy": 0.21194344758987427,
      "num_tokens": 38188816.0,
      "step": 20760
    },
    {
      "entropy": 5.171293306350708,
      "epoch": 1.7344637487470767,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004699231594114349,
      "loss": 4.8567,
      "mean_token_accuracy": 0.21217277497053147,
      "num_tokens": 38198447.0,
      "step": 20765
    },
    {
      "entropy": 5.2080662727355955,
      "epoch": 1.7348813899097895,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004699082881832289,
      "loss": 4.9568,
      "mean_token_accuracy": 0.20604704618453978,
      "num_tokens": 38207075.0,
      "step": 20770
    },
    {
      "entropy": 5.116825532913208,
      "epoch": 1.7352990310725025,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046989341354291864,
      "loss": 4.9204,
      "mean_token_accuracy": 0.2048029661178589,
      "num_tokens": 38216297.0,
      "step": 20775
    },
    {
      "entropy": 5.22699990272522,
      "epoch": 1.7357166722352155,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004698785354907646,
      "loss": 5.0395,
      "mean_token_accuracy": 0.20680301040410995,
      "num_tokens": 38225581.0,
      "step": 20780
    },
    {
      "entropy": 5.095498657226562,
      "epoch": 1.7361343133979283,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004698636540270272,
      "loss": 4.7751,
      "mean_token_accuracy": 0.22054170519113542,
      "num_tokens": 38234214.0,
      "step": 20785
    },
    {
      "entropy": 5.195326566696167,
      "epoch": 1.7365519545606416,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004698487691519671,
      "loss": 4.9893,
      "mean_token_accuracy": 0.1981459692120552,
      "num_tokens": 38243859.0,
      "step": 20790
    },
    {
      "entropy": 5.15996618270874,
      "epoch": 1.7369695957233544,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046983388086584463,
      "loss": 4.8018,
      "mean_token_accuracy": 0.2154707208275795,
      "num_tokens": 38252667.0,
      "step": 20795
    },
    {
      "entropy": 4.973397874832154,
      "epoch": 1.7373872368860674,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046981898916892054,
      "loss": 4.7324,
      "mean_token_accuracy": 0.21574266254901886,
      "num_tokens": 38262216.0,
      "step": 20800
    },
    {
      "entropy": 5.100290727615357,
      "epoch": 1.7378048780487805,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046980409406145555,
      "loss": 4.9093,
      "mean_token_accuracy": 0.20918534994125365,
      "num_tokens": 38271672.0,
      "step": 20805
    },
    {
      "entropy": 5.049693059921265,
      "epoch": 1.7382225192114935,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004697891955437105,
      "loss": 4.8045,
      "mean_token_accuracy": 0.22554027289152145,
      "num_tokens": 38279907.0,
      "step": 20810
    },
    {
      "entropy": 5.079992723464966,
      "epoch": 1.7386401603742065,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00046977429361594596,
      "loss": 4.8019,
      "mean_token_accuracy": 0.22120740860700608,
      "num_tokens": 38289099.0,
      "step": 20815
    },
    {
      "entropy": 5.054329681396484,
      "epoch": 1.7390578015369194,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004697593882784231,
      "loss": 4.8123,
      "mean_token_accuracy": 0.21310171633958816,
      "num_tokens": 38297754.0,
      "step": 20820
    },
    {
      "entropy": 5.165444278717041,
      "epoch": 1.7394754426996326,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004697444795314027,
      "loss": 4.8676,
      "mean_token_accuracy": 0.213735693693161,
      "num_tokens": 38306429.0,
      "step": 20825
    },
    {
      "entropy": 5.147836828231812,
      "epoch": 1.7398930838623454,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046972956737514573,
      "loss": 4.8842,
      "mean_token_accuracy": 0.2150966554880142,
      "num_tokens": 38315505.0,
      "step": 20830
    },
    {
      "entropy": 5.11914963722229,
      "epoch": 1.7403107250250585,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004697146518099133,
      "loss": 4.9054,
      "mean_token_accuracy": 0.21646602153778077,
      "num_tokens": 38324439.0,
      "step": 20835
    },
    {
      "entropy": 5.166195440292358,
      "epoch": 1.7407283661877715,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046969973283596657,
      "loss": 4.8552,
      "mean_token_accuracy": 0.21559954285621644,
      "num_tokens": 38334145.0,
      "step": 20840
    },
    {
      "entropy": 5.1053307056427,
      "epoch": 1.7411460073504843,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046968481045356656,
      "loss": 4.8729,
      "mean_token_accuracy": 0.21596792191267014,
      "num_tokens": 38342811.0,
      "step": 20845
    },
    {
      "entropy": 5.062890434265137,
      "epoch": 1.7415636485131976,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046966988466297464,
      "loss": 4.8708,
      "mean_token_accuracy": 0.2155931442975998,
      "num_tokens": 38351072.0,
      "step": 20850
    },
    {
      "entropy": 5.150931167602539,
      "epoch": 1.7419812896759104,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046965495546445204,
      "loss": 4.8374,
      "mean_token_accuracy": 0.21611802130937577,
      "num_tokens": 38360127.0,
      "step": 20855
    },
    {
      "entropy": 5.111169290542603,
      "epoch": 1.7423989308386234,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046964002285826016,
      "loss": 4.8326,
      "mean_token_accuracy": 0.20905296951532365,
      "num_tokens": 38368270.0,
      "step": 20860
    },
    {
      "entropy": 5.103046131134033,
      "epoch": 1.7428165720013364,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004696250868446604,
      "loss": 4.9316,
      "mean_token_accuracy": 0.19975470006465912,
      "num_tokens": 38376970.0,
      "step": 20865
    },
    {
      "entropy": 5.182408285140991,
      "epoch": 1.7432342131640495,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046961014742391425,
      "loss": 4.9527,
      "mean_token_accuracy": 0.20872612446546554,
      "num_tokens": 38387103.0,
      "step": 20870
    },
    {
      "entropy": 5.240813159942627,
      "epoch": 1.7436518543267625,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046959520459628317,
      "loss": 4.8935,
      "mean_token_accuracy": 0.2088143765926361,
      "num_tokens": 38395890.0,
      "step": 20875
    },
    {
      "entropy": 5.150290489196777,
      "epoch": 1.7440694954894753,
      "grad_norm": 1.390625,
      "learning_rate": 0.00046958025836202886,
      "loss": 4.9702,
      "mean_token_accuracy": 0.20669115483760833,
      "num_tokens": 38405868.0,
      "step": 20880
    },
    {
      "entropy": 5.098982620239258,
      "epoch": 1.7444871366521886,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004695653087214128,
      "loss": 4.9736,
      "mean_token_accuracy": 0.20672862380743026,
      "num_tokens": 38414854.0,
      "step": 20885
    },
    {
      "entropy": 5.09110918045044,
      "epoch": 1.7449047778149014,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046955035567469695,
      "loss": 4.8083,
      "mean_token_accuracy": 0.21347342133522035,
      "num_tokens": 38424158.0,
      "step": 20890
    },
    {
      "entropy": 5.174250173568725,
      "epoch": 1.7453224189776144,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046953539922214285,
      "loss": 4.8834,
      "mean_token_accuracy": 0.21101992577314377,
      "num_tokens": 38433200.0,
      "step": 20895
    },
    {
      "entropy": 5.135856819152832,
      "epoch": 1.7457400601403275,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004695204393640125,
      "loss": 4.8532,
      "mean_token_accuracy": 0.2058097690343857,
      "num_tokens": 38442183.0,
      "step": 20900
    },
    {
      "entropy": 5.120643138885498,
      "epoch": 1.7461577013030403,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046950547610056766,
      "loss": 4.8441,
      "mean_token_accuracy": 0.20444051474332808,
      "num_tokens": 38451451.0,
      "step": 20905
    },
    {
      "entropy": 5.161359643936157,
      "epoch": 1.7465753424657535,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004694905094320703,
      "loss": 4.9012,
      "mean_token_accuracy": 0.20847378820180892,
      "num_tokens": 38460043.0,
      "step": 20910
    },
    {
      "entropy": 5.080163860321045,
      "epoch": 1.7469929836284663,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046947553935878253,
      "loss": 4.9134,
      "mean_token_accuracy": 0.21279775947332383,
      "num_tokens": 38470714.0,
      "step": 20915
    },
    {
      "entropy": 5.178587198257446,
      "epoch": 1.7474106247911794,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004694605658809663,
      "loss": 4.912,
      "mean_token_accuracy": 0.21412128508090972,
      "num_tokens": 38480528.0,
      "step": 20920
    },
    {
      "entropy": 5.095263433456421,
      "epoch": 1.7478282659538924,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004694455889988838,
      "loss": 4.8038,
      "mean_token_accuracy": 0.21140128374099731,
      "num_tokens": 38490101.0,
      "step": 20925
    },
    {
      "entropy": 5.177961444854736,
      "epoch": 1.7482459071166054,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00046943060871279725,
      "loss": 5.0068,
      "mean_token_accuracy": 0.20286845564842224,
      "num_tokens": 38498624.0,
      "step": 20930
    },
    {
      "entropy": 5.105075311660767,
      "epoch": 1.7486635482793185,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046941562502296886,
      "loss": 4.7876,
      "mean_token_accuracy": 0.21714586466550828,
      "num_tokens": 38506646.0,
      "step": 20935
    },
    {
      "entropy": 5.091761493682862,
      "epoch": 1.7490811894420313,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004694006379296609,
      "loss": 4.8092,
      "mean_token_accuracy": 0.2179842010140419,
      "num_tokens": 38516092.0,
      "step": 20940
    },
    {
      "entropy": 5.037694501876831,
      "epoch": 1.7494988306047445,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046938564743313576,
      "loss": 4.8207,
      "mean_token_accuracy": 0.21118590533733367,
      "num_tokens": 38526486.0,
      "step": 20945
    },
    {
      "entropy": 5.0876359939575195,
      "epoch": 1.7499164717674573,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046937065353365594,
      "loss": 4.8374,
      "mean_token_accuracy": 0.21342645734548568,
      "num_tokens": 38535474.0,
      "step": 20950
    },
    {
      "entropy": 5.167647075653076,
      "epoch": 1.7503341129301704,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046935565623148383,
      "loss": 4.8789,
      "mean_token_accuracy": 0.20950808376073837,
      "num_tokens": 38544598.0,
      "step": 20955
    },
    {
      "entropy": 5.132231998443603,
      "epoch": 1.7507517540928834,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046934065552688196,
      "loss": 4.922,
      "mean_token_accuracy": 0.2110055610537529,
      "num_tokens": 38553569.0,
      "step": 20960
    },
    {
      "entropy": 5.160144948959351,
      "epoch": 1.7511693952555962,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004693256514201131,
      "loss": 4.9301,
      "mean_token_accuracy": 0.21069808602333068,
      "num_tokens": 38563369.0,
      "step": 20965
    },
    {
      "entropy": 5.183981657028198,
      "epoch": 1.7515870364183095,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046931064391143976,
      "loss": 4.895,
      "mean_token_accuracy": 0.2070967122912407,
      "num_tokens": 38573187.0,
      "step": 20970
    },
    {
      "entropy": 5.044916725158691,
      "epoch": 1.7520046775810223,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046929563300112466,
      "loss": 4.7457,
      "mean_token_accuracy": 0.2357981562614441,
      "num_tokens": 38581014.0,
      "step": 20975
    },
    {
      "entropy": 5.0550614356994625,
      "epoch": 1.7524223187437353,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046928061868943067,
      "loss": 4.8196,
      "mean_token_accuracy": 0.21262316703796386,
      "num_tokens": 38590445.0,
      "step": 20980
    },
    {
      "entropy": 5.062932252883911,
      "epoch": 1.7528399599064484,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004692656009766206,
      "loss": 4.7985,
      "mean_token_accuracy": 0.2142449513077736,
      "num_tokens": 38599372.0,
      "step": 20985
    },
    {
      "entropy": 5.232157182693482,
      "epoch": 1.7532576010691614,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046925057986295734,
      "loss": 5.0061,
      "mean_token_accuracy": 0.2028690904378891,
      "num_tokens": 38609000.0,
      "step": 20990
    },
    {
      "entropy": 5.122879838943481,
      "epoch": 1.7536752422318744,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046923555534870385,
      "loss": 4.8682,
      "mean_token_accuracy": 0.20809515565633774,
      "num_tokens": 38618420.0,
      "step": 20995
    },
    {
      "entropy": 5.01781816482544,
      "epoch": 1.7540928833945872,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004692205274341232,
      "loss": 4.7701,
      "mean_token_accuracy": 0.22283462285995484,
      "num_tokens": 38627577.0,
      "step": 21000
    },
    {
      "epoch": 1.7540928833945872,
      "eval_entropy": 4.906726895437178,
      "eval_loss": 4.895606517791748,
      "eval_mean_token_accuracy": 0.2211006431493358,
      "eval_num_tokens": 38627577.0,
      "eval_runtime": 27.5156,
      "eval_samples_per_second": 1372.419,
      "eval_steps_per_second": 171.575,
      "step": 21000
    },
    {
      "entropy": 5.187795925140381,
      "epoch": 1.7545105245573005,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046920549611947834,
      "loss": 4.9426,
      "mean_token_accuracy": 0.21254961639642717,
      "num_tokens": 38636837.0,
      "step": 21005
    },
    {
      "entropy": 5.185057926177978,
      "epoch": 1.7549281657200133,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004691904614050326,
      "loss": 4.9204,
      "mean_token_accuracy": 0.2015484869480133,
      "num_tokens": 38646734.0,
      "step": 21010
    },
    {
      "entropy": 5.212862300872803,
      "epoch": 1.7553458068827263,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046917542329104903,
      "loss": 4.9799,
      "mean_token_accuracy": 0.20754479318857194,
      "num_tokens": 38655990.0,
      "step": 21015
    },
    {
      "entropy": 5.126640892028808,
      "epoch": 1.7557634480454394,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004691603817777909,
      "loss": 4.7888,
      "mean_token_accuracy": 0.213624007999897,
      "num_tokens": 38665235.0,
      "step": 21020
    },
    {
      "entropy": 5.099242544174194,
      "epoch": 1.7561810892081522,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046914533686552166,
      "loss": 4.8838,
      "mean_token_accuracy": 0.20539041310548783,
      "num_tokens": 38675810.0,
      "step": 21025
    },
    {
      "entropy": 5.108657693862915,
      "epoch": 1.7565987303708654,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004691302885545046,
      "loss": 4.8286,
      "mean_token_accuracy": 0.212346413731575,
      "num_tokens": 38684702.0,
      "step": 21030
    },
    {
      "entropy": 5.1488560199737545,
      "epoch": 1.7570163715335783,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004691152368450032,
      "loss": 4.9303,
      "mean_token_accuracy": 0.21061402410268784,
      "num_tokens": 38694704.0,
      "step": 21035
    },
    {
      "entropy": 5.086926031112671,
      "epoch": 1.7574340126962913,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046910018173728084,
      "loss": 4.8063,
      "mean_token_accuracy": 0.21240355223417282,
      "num_tokens": 38702674.0,
      "step": 21040
    },
    {
      "entropy": 5.070007085800171,
      "epoch": 1.7578516538590043,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004690851232316012,
      "loss": 4.7873,
      "mean_token_accuracy": 0.22179101407527924,
      "num_tokens": 38711842.0,
      "step": 21045
    },
    {
      "entropy": 5.159300804138184,
      "epoch": 1.7582692950217174,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046907006132822785,
      "loss": 4.9213,
      "mean_token_accuracy": 0.2137123242020607,
      "num_tokens": 38721258.0,
      "step": 21050
    },
    {
      "entropy": 5.086847877502441,
      "epoch": 1.7586869361844304,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046905499602742444,
      "loss": 4.8859,
      "mean_token_accuracy": 0.2137617975473404,
      "num_tokens": 38730170.0,
      "step": 21055
    },
    {
      "entropy": 5.13235502243042,
      "epoch": 1.7591045773471432,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046903992732945485,
      "loss": 5.024,
      "mean_token_accuracy": 0.20633303821086885,
      "num_tokens": 38739790.0,
      "step": 21060
    },
    {
      "entropy": 5.1275012493133545,
      "epoch": 1.7595222185098565,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046902485523458266,
      "loss": 4.8098,
      "mean_token_accuracy": 0.21817754954099655,
      "num_tokens": 38748793.0,
      "step": 21065
    },
    {
      "entropy": 5.135989999771118,
      "epoch": 1.7599398596725693,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046900977974307186,
      "loss": 4.9476,
      "mean_token_accuracy": 0.20766075551509858,
      "num_tokens": 38758067.0,
      "step": 21070
    },
    {
      "entropy": 5.013509225845337,
      "epoch": 1.7603575008352823,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004689947008551863,
      "loss": 4.7901,
      "mean_token_accuracy": 0.2224685311317444,
      "num_tokens": 38767374.0,
      "step": 21075
    },
    {
      "entropy": 5.140185689926147,
      "epoch": 1.7607751419979953,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004689796185711901,
      "loss": 4.9212,
      "mean_token_accuracy": 0.20311895310878753,
      "num_tokens": 38776467.0,
      "step": 21080
    },
    {
      "entropy": 5.156419324874878,
      "epoch": 1.7611927831607082,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046896453289134703,
      "loss": 4.8611,
      "mean_token_accuracy": 0.2149189755320549,
      "num_tokens": 38785916.0,
      "step": 21085
    },
    {
      "entropy": 5.150143575668335,
      "epoch": 1.7616104243234214,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046894944381592143,
      "loss": 4.8915,
      "mean_token_accuracy": 0.21658530086278915,
      "num_tokens": 38795271.0,
      "step": 21090
    },
    {
      "entropy": 5.130131721496582,
      "epoch": 1.7620280654861342,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046893435134517735,
      "loss": 4.9167,
      "mean_token_accuracy": 0.21695092767477037,
      "num_tokens": 38804806.0,
      "step": 21095
    },
    {
      "entropy": 5.092540836334228,
      "epoch": 1.7624457066488475,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046891925547937907,
      "loss": 4.753,
      "mean_token_accuracy": 0.22159518003463746,
      "num_tokens": 38813465.0,
      "step": 21100
    },
    {
      "entropy": 5.076629018783569,
      "epoch": 1.7628633478115603,
      "grad_norm": 1.328125,
      "learning_rate": 0.00046890415621879075,
      "loss": 4.927,
      "mean_token_accuracy": 0.20493336766958237,
      "num_tokens": 38822208.0,
      "step": 21105
    },
    {
      "entropy": 5.172389698028565,
      "epoch": 1.7632809889742733,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004688890535636767,
      "loss": 4.8439,
      "mean_token_accuracy": 0.21714741438627244,
      "num_tokens": 38831136.0,
      "step": 21110
    },
    {
      "entropy": 5.146605062484741,
      "epoch": 1.7636986301369864,
      "grad_norm": 1.375,
      "learning_rate": 0.0004688739475143015,
      "loss": 4.9202,
      "mean_token_accuracy": 0.21135118007659912,
      "num_tokens": 38840995.0,
      "step": 21115
    },
    {
      "entropy": 5.146837425231934,
      "epoch": 1.7641162712996992,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046885883807092944,
      "loss": 4.9634,
      "mean_token_accuracy": 0.21052657812833786,
      "num_tokens": 38850584.0,
      "step": 21120
    },
    {
      "entropy": 5.046890211105347,
      "epoch": 1.7645339124624124,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00046884372523382507,
      "loss": 4.717,
      "mean_token_accuracy": 0.23123919069767,
      "num_tokens": 38858869.0,
      "step": 21125
    },
    {
      "entropy": 5.14423680305481,
      "epoch": 1.7649515536251252,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00046882860900325296,
      "loss": 4.9453,
      "mean_token_accuracy": 0.20132132321596147,
      "num_tokens": 38868701.0,
      "step": 21130
    },
    {
      "entropy": 5.177023887634277,
      "epoch": 1.7653691947878383,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004688134893794777,
      "loss": 4.9972,
      "mean_token_accuracy": 0.20005251616239547,
      "num_tokens": 38877785.0,
      "step": 21135
    },
    {
      "entropy": 5.262628412246704,
      "epoch": 1.7657868359505513,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046879836636276405,
      "loss": 4.9801,
      "mean_token_accuracy": 0.20291673243045807,
      "num_tokens": 38888397.0,
      "step": 21140
    },
    {
      "entropy": 5.1776975154876705,
      "epoch": 1.7662044771132643,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004687832399533767,
      "loss": 4.9583,
      "mean_token_accuracy": 0.20864052027463914,
      "num_tokens": 38897458.0,
      "step": 21145
    },
    {
      "entropy": 5.1900153160095215,
      "epoch": 1.7666221182759774,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004687681101515805,
      "loss": 4.9151,
      "mean_token_accuracy": 0.21068027019500732,
      "num_tokens": 38907536.0,
      "step": 21150
    },
    {
      "entropy": 5.174947214126587,
      "epoch": 1.7670397594386902,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004687529769576403,
      "loss": 4.9288,
      "mean_token_accuracy": 0.21024689823389053,
      "num_tokens": 38917340.0,
      "step": 21155
    },
    {
      "entropy": 5.098480463027954,
      "epoch": 1.7674574006014034,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046873784037182106,
      "loss": 4.9002,
      "mean_token_accuracy": 0.20871923714876175,
      "num_tokens": 38926964.0,
      "step": 21160
    },
    {
      "entropy": 5.085542821884156,
      "epoch": 1.7678750417641163,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046872270039438764,
      "loss": 4.8324,
      "mean_token_accuracy": 0.22082932740449907,
      "num_tokens": 38936168.0,
      "step": 21165
    },
    {
      "entropy": 5.189542961120606,
      "epoch": 1.7682926829268293,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004687075570256052,
      "loss": 5.0181,
      "mean_token_accuracy": 0.20312512218952178,
      "num_tokens": 38945633.0,
      "step": 21170
    },
    {
      "entropy": 5.198152542114258,
      "epoch": 1.7687103240895423,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046869241026573885,
      "loss": 4.9623,
      "mean_token_accuracy": 0.20431272089481353,
      "num_tokens": 38955372.0,
      "step": 21175
    },
    {
      "entropy": 5.164313268661499,
      "epoch": 1.7691279652522551,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046867726011505365,
      "loss": 4.8702,
      "mean_token_accuracy": 0.21060038208961487,
      "num_tokens": 38965264.0,
      "step": 21180
    },
    {
      "entropy": 5.138087940216065,
      "epoch": 1.7695456064149684,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004686621065738149,
      "loss": 4.8989,
      "mean_token_accuracy": 0.2148103103041649,
      "num_tokens": 38975129.0,
      "step": 21185
    },
    {
      "entropy": 5.192765760421753,
      "epoch": 1.7699632475776812,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046864694964228785,
      "loss": 5.0073,
      "mean_token_accuracy": 0.20582732409238816,
      "num_tokens": 38985152.0,
      "step": 21190
    },
    {
      "entropy": 5.068816900253296,
      "epoch": 1.7703808887403942,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004686317893207378,
      "loss": 4.8128,
      "mean_token_accuracy": 0.20785613656044005,
      "num_tokens": 38995509.0,
      "step": 21195
    },
    {
      "entropy": 5.061005973815918,
      "epoch": 1.7707985299031073,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004686166256094303,
      "loss": 4.8745,
      "mean_token_accuracy": 0.20727640986442566,
      "num_tokens": 39003743.0,
      "step": 21200
    },
    {
      "entropy": 5.164299201965332,
      "epoch": 1.7712161710658203,
      "grad_norm": 1.125,
      "learning_rate": 0.0004686014585086307,
      "loss": 4.9101,
      "mean_token_accuracy": 0.20954231172800064,
      "num_tokens": 39013324.0,
      "step": 21205
    },
    {
      "entropy": 5.10254921913147,
      "epoch": 1.7716338122285333,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004685862880186045,
      "loss": 4.8709,
      "mean_token_accuracy": 0.21447789520025254,
      "num_tokens": 39023268.0,
      "step": 21210
    },
    {
      "entropy": 5.155828332901001,
      "epoch": 1.7720514533912461,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004685711141396173,
      "loss": 4.9427,
      "mean_token_accuracy": 0.20875197947025298,
      "num_tokens": 39033516.0,
      "step": 21215
    },
    {
      "entropy": 5.156027126312256,
      "epoch": 1.7724690945539594,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00046855593687193474,
      "loss": 4.9367,
      "mean_token_accuracy": 0.20827065259218216,
      "num_tokens": 39042280.0,
      "step": 21220
    },
    {
      "entropy": 5.054925012588501,
      "epoch": 1.7728867357166722,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004685407562158226,
      "loss": 4.891,
      "mean_token_accuracy": 0.2219451755285263,
      "num_tokens": 39051748.0,
      "step": 21225
    },
    {
      "entropy": 5.183876705169678,
      "epoch": 1.7733043768793852,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004685255721715465,
      "loss": 4.9316,
      "mean_token_accuracy": 0.20395064651966094,
      "num_tokens": 39061703.0,
      "step": 21230
    },
    {
      "entropy": 5.221156740188599,
      "epoch": 1.7737220180420983,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004685103847393723,
      "loss": 4.9294,
      "mean_token_accuracy": 0.20924082249403,
      "num_tokens": 39071115.0,
      "step": 21235
    },
    {
      "entropy": 5.125373506546021,
      "epoch": 1.774139659204811,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046849519391956593,
      "loss": 4.8905,
      "mean_token_accuracy": 0.20417936742305756,
      "num_tokens": 39080578.0,
      "step": 21240
    },
    {
      "entropy": 5.103058576583862,
      "epoch": 1.7745573003675243,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004684799997123932,
      "loss": 4.9121,
      "mean_token_accuracy": 0.20994067937135696,
      "num_tokens": 39090242.0,
      "step": 21245
    },
    {
      "entropy": 5.184209537506104,
      "epoch": 1.7749749415302372,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004684648021181203,
      "loss": 4.9389,
      "mean_token_accuracy": 0.20979838222265243,
      "num_tokens": 39099831.0,
      "step": 21250
    },
    {
      "entropy": 5.052172899246216,
      "epoch": 1.7753925826929502,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046844960113701307,
      "loss": 4.7245,
      "mean_token_accuracy": 0.2218891993165016,
      "num_tokens": 39109047.0,
      "step": 21255
    },
    {
      "entropy": 5.0661702156066895,
      "epoch": 1.7758102238556632,
      "grad_norm": 1.359375,
      "learning_rate": 0.00046843439676933776,
      "loss": 4.8301,
      "mean_token_accuracy": 0.2151439070701599,
      "num_tokens": 39117800.0,
      "step": 21260
    },
    {
      "entropy": 5.150977373123169,
      "epoch": 1.7762278650183763,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046841918901536047,
      "loss": 4.921,
      "mean_token_accuracy": 0.21000135242938994,
      "num_tokens": 39127019.0,
      "step": 21265
    },
    {
      "entropy": 5.0799839973449705,
      "epoch": 1.7766455061810893,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004684039778753474,
      "loss": 4.7779,
      "mean_token_accuracy": 0.22258027046918868,
      "num_tokens": 39135983.0,
      "step": 21270
    },
    {
      "entropy": 5.109597682952881,
      "epoch": 1.777063147343802,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046838876334956507,
      "loss": 4.8259,
      "mean_token_accuracy": 0.21047288477420806,
      "num_tokens": 39145890.0,
      "step": 21275
    },
    {
      "entropy": 5.1071454048156735,
      "epoch": 1.7774807885065154,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004683735454382795,
      "loss": 4.8137,
      "mean_token_accuracy": 0.2095888525247574,
      "num_tokens": 39154884.0,
      "step": 21280
    },
    {
      "entropy": 5.120285749435425,
      "epoch": 1.7778984296692282,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004683583241417573,
      "loss": 4.8555,
      "mean_token_accuracy": 0.20898054391145707,
      "num_tokens": 39164151.0,
      "step": 21285
    },
    {
      "entropy": 5.202142143249512,
      "epoch": 1.7783160708319412,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004683430994602649,
      "loss": 4.9852,
      "mean_token_accuracy": 0.20337690860033036,
      "num_tokens": 39173987.0,
      "step": 21290
    },
    {
      "entropy": 5.1459191799163815,
      "epoch": 1.7787337119946542,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046832787139406887,
      "loss": 4.9832,
      "mean_token_accuracy": 0.21366076022386551,
      "num_tokens": 39183665.0,
      "step": 21295
    },
    {
      "entropy": 5.123756265640258,
      "epoch": 1.779151353157367,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004683126399434357,
      "loss": 4.8281,
      "mean_token_accuracy": 0.217282797396183,
      "num_tokens": 39192897.0,
      "step": 21300
    },
    {
      "entropy": 5.223187589645386,
      "epoch": 1.7795689943200803,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004682974051086321,
      "loss": 4.9945,
      "mean_token_accuracy": 0.2063418447971344,
      "num_tokens": 39202774.0,
      "step": 21305
    },
    {
      "entropy": 5.108062267303467,
      "epoch": 1.7799866354827931,
      "grad_norm": 1.25,
      "learning_rate": 0.00046828216688992466,
      "loss": 4.8438,
      "mean_token_accuracy": 0.21076417565345765,
      "num_tokens": 39212154.0,
      "step": 21310
    },
    {
      "entropy": 5.042844200134278,
      "epoch": 1.7804042766455062,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004682669252875803,
      "loss": 4.781,
      "mean_token_accuracy": 0.21802304387092591,
      "num_tokens": 39221593.0,
      "step": 21315
    },
    {
      "entropy": 5.129908084869385,
      "epoch": 1.7808219178082192,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046825168030186577,
      "loss": 4.9004,
      "mean_token_accuracy": 0.20956984907388687,
      "num_tokens": 39230522.0,
      "step": 21320
    },
    {
      "entropy": 5.167433452606201,
      "epoch": 1.7812395589709322,
      "grad_norm": 1.234375,
      "learning_rate": 0.000468236431933048,
      "loss": 4.9391,
      "mean_token_accuracy": 0.21071948111057281,
      "num_tokens": 39240012.0,
      "step": 21325
    },
    {
      "entropy": 5.127293968200684,
      "epoch": 1.7816572001336453,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004682211801813938,
      "loss": 4.8432,
      "mean_token_accuracy": 0.21511461436748505,
      "num_tokens": 39248789.0,
      "step": 21330
    },
    {
      "entropy": 5.207061338424682,
      "epoch": 1.782074841296358,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004682059250471703,
      "loss": 4.912,
      "mean_token_accuracy": 0.2144966185092926,
      "num_tokens": 39257864.0,
      "step": 21335
    },
    {
      "entropy": 5.192798137664795,
      "epoch": 1.7824924824590713,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046819066653064454,
      "loss": 4.8382,
      "mean_token_accuracy": 0.20795388370752335,
      "num_tokens": 39267137.0,
      "step": 21340
    },
    {
      "entropy": 5.038035535812378,
      "epoch": 1.7829101236217841,
      "grad_norm": 1.375,
      "learning_rate": 0.00046817540463208357,
      "loss": 4.9323,
      "mean_token_accuracy": 0.2097514435648918,
      "num_tokens": 39276511.0,
      "step": 21345
    },
    {
      "entropy": 5.083869409561157,
      "epoch": 1.7833277647844972,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046816013935175465,
      "loss": 4.8428,
      "mean_token_accuracy": 0.22212165892124175,
      "num_tokens": 39285053.0,
      "step": 21350
    },
    {
      "entropy": 5.081965208053589,
      "epoch": 1.7837454059472102,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000468144870689925,
      "loss": 4.8302,
      "mean_token_accuracy": 0.21405524462461473,
      "num_tokens": 39293536.0,
      "step": 21355
    },
    {
      "entropy": 5.206602144241333,
      "epoch": 1.784163047109923,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004681295986468619,
      "loss": 4.9397,
      "mean_token_accuracy": 0.203043070435524,
      "num_tokens": 39303074.0,
      "step": 21360
    },
    {
      "entropy": 5.141311931610107,
      "epoch": 1.7845806882726363,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004681143232228328,
      "loss": 4.8748,
      "mean_token_accuracy": 0.20985658317804337,
      "num_tokens": 39312148.0,
      "step": 21365
    },
    {
      "entropy": 5.15810604095459,
      "epoch": 1.784998329435349,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046809904441810486,
      "loss": 4.8964,
      "mean_token_accuracy": 0.21404281556606292,
      "num_tokens": 39321098.0,
      "step": 21370
    },
    {
      "entropy": 5.050575637817383,
      "epoch": 1.7854159705980621,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004680837622329457,
      "loss": 4.705,
      "mean_token_accuracy": 0.2194462388753891,
      "num_tokens": 39329463.0,
      "step": 21375
    },
    {
      "entropy": 5.092411851882934,
      "epoch": 1.7858336117607752,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046806847666762305,
      "loss": 4.8693,
      "mean_token_accuracy": 0.21470358073711396,
      "num_tokens": 39339497.0,
      "step": 21380
    },
    {
      "entropy": 5.061924695968628,
      "epoch": 1.7862512529234882,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004680531877224042,
      "loss": 4.7425,
      "mean_token_accuracy": 0.2216583088040352,
      "num_tokens": 39348121.0,
      "step": 21385
    },
    {
      "entropy": 5.087414312362671,
      "epoch": 1.7866688940862012,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046803789539755686,
      "loss": 4.9301,
      "mean_token_accuracy": 0.20992299020290375,
      "num_tokens": 39358487.0,
      "step": 21390
    },
    {
      "entropy": 5.102935123443603,
      "epoch": 1.787086535248914,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046802259969334887,
      "loss": 4.9162,
      "mean_token_accuracy": 0.20958223640918733,
      "num_tokens": 39367493.0,
      "step": 21395
    },
    {
      "entropy": 5.08837857246399,
      "epoch": 1.7875041764116273,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004680073006100479,
      "loss": 4.857,
      "mean_token_accuracy": 0.21090839058160782,
      "num_tokens": 39376616.0,
      "step": 21400
    },
    {
      "entropy": 5.188196897506714,
      "epoch": 1.78792181757434,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046799199814792186,
      "loss": 4.9462,
      "mean_token_accuracy": 0.20642078667879105,
      "num_tokens": 39387079.0,
      "step": 21405
    },
    {
      "entropy": 5.0318458557128904,
      "epoch": 1.7883394587370531,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004679766923072386,
      "loss": 4.7497,
      "mean_token_accuracy": 0.22225138396024705,
      "num_tokens": 39396198.0,
      "step": 21410
    },
    {
      "entropy": 5.177270555496216,
      "epoch": 1.7887570998997662,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000467961383088266,
      "loss": 4.9848,
      "mean_token_accuracy": 0.20699334889650345,
      "num_tokens": 39405862.0,
      "step": 21415
    },
    {
      "entropy": 5.154102611541748,
      "epoch": 1.789174741062479,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046794607049127217,
      "loss": 4.9051,
      "mean_token_accuracy": 0.2078057199716568,
      "num_tokens": 39414352.0,
      "step": 21420
    },
    {
      "entropy": 5.182077646255493,
      "epoch": 1.7895923822251922,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046793075451652513,
      "loss": 4.9376,
      "mean_token_accuracy": 0.20376965701580046,
      "num_tokens": 39423653.0,
      "step": 21425
    },
    {
      "entropy": 5.209628057479859,
      "epoch": 1.790010023387905,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046791543516429295,
      "loss": 4.9378,
      "mean_token_accuracy": 0.2120342254638672,
      "num_tokens": 39432741.0,
      "step": 21430
    },
    {
      "entropy": 5.229291391372681,
      "epoch": 1.790427664550618,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046790011243484387,
      "loss": 4.9192,
      "mean_token_accuracy": 0.20979571342468262,
      "num_tokens": 39441837.0,
      "step": 21435
    },
    {
      "entropy": 5.087250185012818,
      "epoch": 1.7908453057133311,
      "grad_norm": 1.375,
      "learning_rate": 0.00046788478632844616,
      "loss": 4.8513,
      "mean_token_accuracy": 0.21417894810438157,
      "num_tokens": 39450393.0,
      "step": 21440
    },
    {
      "entropy": 5.066080093383789,
      "epoch": 1.7912629468760441,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046786945684536806,
      "loss": 4.7966,
      "mean_token_accuracy": 0.22214146405458451,
      "num_tokens": 39460226.0,
      "step": 21445
    },
    {
      "entropy": 5.1027978420257565,
      "epoch": 1.7916805880387572,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000467854123985878,
      "loss": 4.7835,
      "mean_token_accuracy": 0.2144453376531601,
      "num_tokens": 39469072.0,
      "step": 21450
    },
    {
      "entropy": 5.103954553604126,
      "epoch": 1.79209822920147,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004678387877502443,
      "loss": 4.8345,
      "mean_token_accuracy": 0.21180603951215743,
      "num_tokens": 39477918.0,
      "step": 21455
    },
    {
      "entropy": 4.949414539337158,
      "epoch": 1.7925158703641833,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046782344813873556,
      "loss": 4.7375,
      "mean_token_accuracy": 0.22470580339431762,
      "num_tokens": 39487492.0,
      "step": 21460
    },
    {
      "entropy": 5.073260164260864,
      "epoch": 1.792933511526896,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046780810515162027,
      "loss": 4.828,
      "mean_token_accuracy": 0.22323760241270066,
      "num_tokens": 39496255.0,
      "step": 21465
    },
    {
      "entropy": 5.036720037460327,
      "epoch": 1.793351152689609,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00046779275878916694,
      "loss": 4.8167,
      "mean_token_accuracy": 0.22376493960618973,
      "num_tokens": 39504925.0,
      "step": 21470
    },
    {
      "entropy": 5.125568389892578,
      "epoch": 1.7937687938523221,
      "grad_norm": 1.328125,
      "learning_rate": 0.00046777740905164434,
      "loss": 4.9117,
      "mean_token_accuracy": 0.20533362329006194,
      "num_tokens": 39514289.0,
      "step": 21475
    },
    {
      "entropy": 5.201919984817505,
      "epoch": 1.794186435015035,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046776205593932114,
      "loss": 4.9275,
      "mean_token_accuracy": 0.2044031649827957,
      "num_tokens": 39524421.0,
      "step": 21480
    },
    {
      "entropy": 5.184623670578003,
      "epoch": 1.7946040761777482,
      "grad_norm": 1.265625,
      "learning_rate": 0.00046774669945246616,
      "loss": 4.8257,
      "mean_token_accuracy": 0.21112625449895858,
      "num_tokens": 39532866.0,
      "step": 21485
    },
    {
      "entropy": 5.151296949386596,
      "epoch": 1.795021717340461,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004677313395913482,
      "loss": 4.9393,
      "mean_token_accuracy": 0.2053874522447586,
      "num_tokens": 39542964.0,
      "step": 21490
    },
    {
      "entropy": 5.186654853820801,
      "epoch": 1.795439358503174,
      "grad_norm": 1.25,
      "learning_rate": 0.0004677159763562361,
      "loss": 4.945,
      "mean_token_accuracy": 0.20367584526538848,
      "num_tokens": 39554127.0,
      "step": 21495
    },
    {
      "entropy": 5.126865291595459,
      "epoch": 1.795856999665887,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046770060974739893,
      "loss": 4.8513,
      "mean_token_accuracy": 0.2136510953307152,
      "num_tokens": 39563600.0,
      "step": 21500
    },
    {
      "entropy": 5.052946710586548,
      "epoch": 1.7962746408286,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004676852397651055,
      "loss": 4.819,
      "mean_token_accuracy": 0.21101032793521882,
      "num_tokens": 39572502.0,
      "step": 21505
    },
    {
      "entropy": 5.115253067016601,
      "epoch": 1.7966922819913131,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004676698664096251,
      "loss": 4.7539,
      "mean_token_accuracy": 0.219573450088501,
      "num_tokens": 39581066.0,
      "step": 21510
    },
    {
      "entropy": 5.081646251678467,
      "epoch": 1.797109923154026,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004676544896812268,
      "loss": 4.8335,
      "mean_token_accuracy": 0.2175368309020996,
      "num_tokens": 39589784.0,
      "step": 21515
    },
    {
      "entropy": 5.1113365650177,
      "epoch": 1.7975275643167392,
      "grad_norm": 1.34375,
      "learning_rate": 0.00046763910958017973,
      "loss": 4.9662,
      "mean_token_accuracy": 0.20241871923208238,
      "num_tokens": 39598561.0,
      "step": 21520
    },
    {
      "entropy": 5.1193911075592045,
      "epoch": 1.797945205479452,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046762372610675314,
      "loss": 4.7717,
      "mean_token_accuracy": 0.21888373494148256,
      "num_tokens": 39607136.0,
      "step": 21525
    },
    {
      "entropy": 5.183148670196533,
      "epoch": 1.798362846642165,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00046760833926121645,
      "loss": 4.8967,
      "mean_token_accuracy": 0.2151231810450554,
      "num_tokens": 39616371.0,
      "step": 21530
    },
    {
      "entropy": 5.0774884700775145,
      "epoch": 1.798780487804878,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046759294904383887,
      "loss": 4.8821,
      "mean_token_accuracy": 0.21709989458322526,
      "num_tokens": 39625556.0,
      "step": 21535
    },
    {
      "entropy": 5.090655040740967,
      "epoch": 1.799198128967591,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004675775554548899,
      "loss": 4.8135,
      "mean_token_accuracy": 0.2121976748108864,
      "num_tokens": 39635116.0,
      "step": 21540
    },
    {
      "entropy": 5.170913887023926,
      "epoch": 1.7996157701303042,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00046756215849463905,
      "loss": 4.8946,
      "mean_token_accuracy": 0.20742181241512297,
      "num_tokens": 39645523.0,
      "step": 21545
    },
    {
      "entropy": 5.173139238357544,
      "epoch": 1.800033411293017,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00046754675816335577,
      "loss": 4.9794,
      "mean_token_accuracy": 0.20139268934726715,
      "num_tokens": 39654507.0,
      "step": 21550
    },
    {
      "entropy": 5.124932384490966,
      "epoch": 1.80045105245573,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046753135446130973,
      "loss": 4.8538,
      "mean_token_accuracy": 0.21614550650119782,
      "num_tokens": 39664587.0,
      "step": 21555
    },
    {
      "entropy": 5.160187101364135,
      "epoch": 1.800868693618443,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004675159473887707,
      "loss": 4.9802,
      "mean_token_accuracy": 0.20386249870061873,
      "num_tokens": 39674844.0,
      "step": 21560
    },
    {
      "entropy": 5.073069715499878,
      "epoch": 1.801286334781156,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004675005369460081,
      "loss": 4.8475,
      "mean_token_accuracy": 0.2156723454594612,
      "num_tokens": 39683894.0,
      "step": 21565
    },
    {
      "entropy": 5.094110631942749,
      "epoch": 1.801703975943869,
      "grad_norm": 1.234375,
      "learning_rate": 0.000467485123133292,
      "loss": 4.9147,
      "mean_token_accuracy": 0.21064065545797347,
      "num_tokens": 39693401.0,
      "step": 21570
    },
    {
      "entropy": 5.148292350769043,
      "epoch": 1.802121617106582,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00046746970595089207,
      "loss": 4.9118,
      "mean_token_accuracy": 0.20527471005916595,
      "num_tokens": 39702506.0,
      "step": 21575
    },
    {
      "entropy": 5.1354694843292235,
      "epoch": 1.8025392582692952,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004674542853990783,
      "loss": 4.8348,
      "mean_token_accuracy": 0.21191304177045822,
      "num_tokens": 39711252.0,
      "step": 21580
    },
    {
      "entropy": 5.113564395904541,
      "epoch": 1.802956899432008,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046743886147812054,
      "loss": 4.8512,
      "mean_token_accuracy": 0.21364850997924806,
      "num_tokens": 39719803.0,
      "step": 21585
    },
    {
      "entropy": 5.066733646392822,
      "epoch": 1.803374540594721,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004674234341882889,
      "loss": 4.7437,
      "mean_token_accuracy": 0.2279031440615654,
      "num_tokens": 39728492.0,
      "step": 21590
    },
    {
      "entropy": 5.112820434570312,
      "epoch": 1.803792181757434,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004674080035298535,
      "loss": 4.945,
      "mean_token_accuracy": 0.20059346854686738,
      "num_tokens": 39737703.0,
      "step": 21595
    },
    {
      "entropy": 5.108752155303955,
      "epoch": 1.8042098229201469,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004673925695030843,
      "loss": 4.852,
      "mean_token_accuracy": 0.2105630487203598,
      "num_tokens": 39746529.0,
      "step": 21600
    },
    {
      "entropy": 5.23503303527832,
      "epoch": 1.8046274640828601,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004673771321082517,
      "loss": 5.0568,
      "mean_token_accuracy": 0.19896549433469773,
      "num_tokens": 39756398.0,
      "step": 21605
    },
    {
      "entropy": 5.199801683425903,
      "epoch": 1.805045105245573,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004673616913456258,
      "loss": 4.9526,
      "mean_token_accuracy": 0.20392027497291565,
      "num_tokens": 39766354.0,
      "step": 21610
    },
    {
      "entropy": 5.201670837402344,
      "epoch": 1.805462746408286,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046734624721547687,
      "loss": 5.0001,
      "mean_token_accuracy": 0.19889099448919295,
      "num_tokens": 39776089.0,
      "step": 21615
    },
    {
      "entropy": 5.127684736251831,
      "epoch": 1.805880387570999,
      "grad_norm": 1.25,
      "learning_rate": 0.00046733079971807536,
      "loss": 4.9107,
      "mean_token_accuracy": 0.2113633409142494,
      "num_tokens": 39785390.0,
      "step": 21620
    },
    {
      "entropy": 5.065797185897827,
      "epoch": 1.806298028733712,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046731534885369175,
      "loss": 4.6874,
      "mean_token_accuracy": 0.22587241530418395,
      "num_tokens": 39793837.0,
      "step": 21625
    },
    {
      "entropy": 5.111856079101562,
      "epoch": 1.806715669896425,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004672998946225965,
      "loss": 4.8707,
      "mean_token_accuracy": 0.20975372493267058,
      "num_tokens": 39802391.0,
      "step": 21630
    },
    {
      "entropy": 5.061371326446533,
      "epoch": 1.8071333110591379,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046728443702506007,
      "loss": 4.8632,
      "mean_token_accuracy": 0.2098311588168144,
      "num_tokens": 39812110.0,
      "step": 21635
    },
    {
      "entropy": 5.146898984909058,
      "epoch": 1.8075509522218511,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046726897606135307,
      "loss": 4.8747,
      "mean_token_accuracy": 0.20792758464813232,
      "num_tokens": 39821647.0,
      "step": 21640
    },
    {
      "entropy": 5.074562406539917,
      "epoch": 1.807968593384564,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046725351173174627,
      "loss": 4.7998,
      "mean_token_accuracy": 0.21864552646875382,
      "num_tokens": 39829520.0,
      "step": 21645
    },
    {
      "entropy": 5.115319728851318,
      "epoch": 1.808386234547277,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004672380440365103,
      "loss": 4.9496,
      "mean_token_accuracy": 0.2085807204246521,
      "num_tokens": 39839226.0,
      "step": 21650
    },
    {
      "entropy": 5.171886730194092,
      "epoch": 1.80880387570999,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000467222572975916,
      "loss": 4.8787,
      "mean_token_accuracy": 0.22290316671133042,
      "num_tokens": 39847937.0,
      "step": 21655
    },
    {
      "entropy": 5.0793016910552975,
      "epoch": 1.8092215168727028,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004672070985502342,
      "loss": 4.8029,
      "mean_token_accuracy": 0.21484173387289046,
      "num_tokens": 39857293.0,
      "step": 21660
    },
    {
      "entropy": 5.024575805664062,
      "epoch": 1.809639158035416,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046719162075973576,
      "loss": 4.7818,
      "mean_token_accuracy": 0.21750315874814988,
      "num_tokens": 39867058.0,
      "step": 21665
    },
    {
      "entropy": 4.993581008911133,
      "epoch": 1.810056799198129,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004671761396046916,
      "loss": 4.793,
      "mean_token_accuracy": 0.22206286042928697,
      "num_tokens": 39876166.0,
      "step": 21670
    },
    {
      "entropy": 5.104401302337647,
      "epoch": 1.810474440360842,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004671606550853729,
      "loss": 4.9291,
      "mean_token_accuracy": 0.20816198885440826,
      "num_tokens": 39885980.0,
      "step": 21675
    },
    {
      "entropy": 5.181935548782349,
      "epoch": 1.810892081523555,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046714516720205057,
      "loss": 4.972,
      "mean_token_accuracy": 0.20673005431890487,
      "num_tokens": 39895263.0,
      "step": 21680
    },
    {
      "entropy": 5.134307193756103,
      "epoch": 1.811309722686268,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046712967595499585,
      "loss": 4.974,
      "mean_token_accuracy": 0.20239317417144775,
      "num_tokens": 39904052.0,
      "step": 21685
    },
    {
      "entropy": 5.059835815429688,
      "epoch": 1.811727363848981,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004671141813444798,
      "loss": 4.8274,
      "mean_token_accuracy": 0.21839730739593505,
      "num_tokens": 39913411.0,
      "step": 21690
    },
    {
      "entropy": 5.103522825241089,
      "epoch": 1.8121450050116938,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046709868337077386,
      "loss": 4.7647,
      "mean_token_accuracy": 0.21990558505058289,
      "num_tokens": 39922262.0,
      "step": 21695
    },
    {
      "entropy": 5.150766038894654,
      "epoch": 1.812562646174407,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004670831820341491,
      "loss": 5.0028,
      "mean_token_accuracy": 0.20669835954904556,
      "num_tokens": 39932067.0,
      "step": 21700
    },
    {
      "entropy": 5.176459884643554,
      "epoch": 1.81298028733712,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046706767733487714,
      "loss": 4.9247,
      "mean_token_accuracy": 0.21071914881467818,
      "num_tokens": 39941565.0,
      "step": 21705
    },
    {
      "entropy": 5.094394969940185,
      "epoch": 1.813397928499833,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046705216927322925,
      "loss": 4.7431,
      "mean_token_accuracy": 0.21977942436933517,
      "num_tokens": 39950569.0,
      "step": 21710
    },
    {
      "entropy": 5.056819295883178,
      "epoch": 1.813815569662546,
      "grad_norm": 1.125,
      "learning_rate": 0.000467036657849477,
      "loss": 4.7936,
      "mean_token_accuracy": 0.2181509643793106,
      "num_tokens": 39959341.0,
      "step": 21715
    },
    {
      "entropy": 5.205641412734986,
      "epoch": 1.8142332108252588,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004670211430638919,
      "loss": 4.9813,
      "mean_token_accuracy": 0.20369830280542373,
      "num_tokens": 39968503.0,
      "step": 21720
    },
    {
      "entropy": 5.1633415699005125,
      "epoch": 1.814650851987972,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004670056249167455,
      "loss": 4.9658,
      "mean_token_accuracy": 0.19755005836486816,
      "num_tokens": 39977717.0,
      "step": 21725
    },
    {
      "entropy": 5.020628452301025,
      "epoch": 1.8150684931506849,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004669901034083096,
      "loss": 4.8076,
      "mean_token_accuracy": 0.20759740620851516,
      "num_tokens": 39986066.0,
      "step": 21730
    },
    {
      "entropy": 5.201438331604004,
      "epoch": 1.815486134313398,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046697457853885576,
      "loss": 4.9602,
      "mean_token_accuracy": 0.20082937479019164,
      "num_tokens": 39995067.0,
      "step": 21735
    },
    {
      "entropy": 5.156808090209961,
      "epoch": 1.815903775476111,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004669590503086559,
      "loss": 4.8782,
      "mean_token_accuracy": 0.21192185580730438,
      "num_tokens": 40005102.0,
      "step": 21740
    },
    {
      "entropy": 5.09385347366333,
      "epoch": 1.816321416638824,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046694351871798175,
      "loss": 4.7997,
      "mean_token_accuracy": 0.21436715126037598,
      "num_tokens": 40013475.0,
      "step": 21745
    },
    {
      "entropy": 5.129926919937134,
      "epoch": 1.816739057801537,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00046692798376710533,
      "loss": 4.8955,
      "mean_token_accuracy": 0.21004850715398787,
      "num_tokens": 40023083.0,
      "step": 21750
    },
    {
      "entropy": 4.95221209526062,
      "epoch": 1.8171566989642498,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046691244545629847,
      "loss": 4.7748,
      "mean_token_accuracy": 0.22232675552368164,
      "num_tokens": 40031462.0,
      "step": 21755
    },
    {
      "entropy": 5.153171062469482,
      "epoch": 1.817574340126963,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004668969037858333,
      "loss": 4.8757,
      "mean_token_accuracy": 0.20685852319002151,
      "num_tokens": 40040037.0,
      "step": 21760
    },
    {
      "entropy": 5.212763166427612,
      "epoch": 1.8179919812896759,
      "grad_norm": 1.390625,
      "learning_rate": 0.00046688135875598184,
      "loss": 4.8896,
      "mean_token_accuracy": 0.20552038699388503,
      "num_tokens": 40049668.0,
      "step": 21765
    },
    {
      "entropy": 5.128484869003296,
      "epoch": 1.818409622452389,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004668658103670161,
      "loss": 4.8613,
      "mean_token_accuracy": 0.2098556712269783,
      "num_tokens": 40058080.0,
      "step": 21770
    },
    {
      "entropy": 5.185696458816528,
      "epoch": 1.818827263615102,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046685025861920856,
      "loss": 4.9405,
      "mean_token_accuracy": 0.2156794086098671,
      "num_tokens": 40067411.0,
      "step": 21775
    },
    {
      "entropy": 5.089353132247925,
      "epoch": 1.8192449047778148,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004668347035128312,
      "loss": 4.8346,
      "mean_token_accuracy": 0.20959800332784653,
      "num_tokens": 40076481.0,
      "step": 21780
    },
    {
      "entropy": 5.1034139633178714,
      "epoch": 1.819662545940528,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046681914504815653,
      "loss": 4.9939,
      "mean_token_accuracy": 0.20930842757225038,
      "num_tokens": 40084981.0,
      "step": 21785
    },
    {
      "entropy": 5.149927282333374,
      "epoch": 1.8200801871032408,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046680358322545687,
      "loss": 4.8602,
      "mean_token_accuracy": 0.21626912355422973,
      "num_tokens": 40094219.0,
      "step": 21790
    },
    {
      "entropy": 5.168687582015991,
      "epoch": 1.8204978282659539,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004667880180450045,
      "loss": 4.8865,
      "mean_token_accuracy": 0.21369108706712722,
      "num_tokens": 40103833.0,
      "step": 21795
    },
    {
      "entropy": 5.156848526000976,
      "epoch": 1.8209154694286669,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046677244950707204,
      "loss": 4.9154,
      "mean_token_accuracy": 0.20635705888271333,
      "num_tokens": 40112906.0,
      "step": 21800
    },
    {
      "entropy": 5.109606456756592,
      "epoch": 1.82133311059138,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004667568776119321,
      "loss": 4.8999,
      "mean_token_accuracy": 0.20399967730045318,
      "num_tokens": 40122080.0,
      "step": 21805
    },
    {
      "entropy": 5.11511492729187,
      "epoch": 1.821750751754093,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004667413023598571,
      "loss": 4.866,
      "mean_token_accuracy": 0.21485136747360228,
      "num_tokens": 40131765.0,
      "step": 21810
    },
    {
      "entropy": 5.066268968582153,
      "epoch": 1.8221683929168058,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004667257237511199,
      "loss": 4.7926,
      "mean_token_accuracy": 0.2163222387433052,
      "num_tokens": 40141084.0,
      "step": 21815
    },
    {
      "entropy": 5.0794572830200195,
      "epoch": 1.822586034079519,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046671014178599295,
      "loss": 4.8251,
      "mean_token_accuracy": 0.2142500624060631,
      "num_tokens": 40149845.0,
      "step": 21820
    },
    {
      "entropy": 5.143493700027466,
      "epoch": 1.8230036752422318,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046669455646474936,
      "loss": 4.8678,
      "mean_token_accuracy": 0.20837997049093246,
      "num_tokens": 40159173.0,
      "step": 21825
    },
    {
      "entropy": 5.129666328430176,
      "epoch": 1.8234213164049449,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046667896778766185,
      "loss": 4.932,
      "mean_token_accuracy": 0.20779924392700194,
      "num_tokens": 40168949.0,
      "step": 21830
    },
    {
      "entropy": 5.099366188049316,
      "epoch": 1.823838957567658,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046666337575500315,
      "loss": 4.7985,
      "mean_token_accuracy": 0.20975281149148942,
      "num_tokens": 40178886.0,
      "step": 21835
    },
    {
      "entropy": 5.114319515228272,
      "epoch": 1.8242565987303707,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004666477803670464,
      "loss": 4.8171,
      "mean_token_accuracy": 0.21294882297515869,
      "num_tokens": 40187562.0,
      "step": 21840
    },
    {
      "entropy": 5.11909441947937,
      "epoch": 1.824674239893084,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004666321816240646,
      "loss": 4.8662,
      "mean_token_accuracy": 0.22136213183403014,
      "num_tokens": 40196910.0,
      "step": 21845
    },
    {
      "entropy": 5.085117959976197,
      "epoch": 1.8250918810557968,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004666165795263308,
      "loss": 4.8865,
      "mean_token_accuracy": 0.21518223136663436,
      "num_tokens": 40206238.0,
      "step": 21850
    },
    {
      "entropy": 5.100681304931641,
      "epoch": 1.8255095222185098,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004666009740741181,
      "loss": 4.8674,
      "mean_token_accuracy": 0.2193135216832161,
      "num_tokens": 40215191.0,
      "step": 21855
    },
    {
      "entropy": 5.11305251121521,
      "epoch": 1.8259271633812229,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046658536526769976,
      "loss": 4.8717,
      "mean_token_accuracy": 0.21617553234100342,
      "num_tokens": 40224408.0,
      "step": 21860
    },
    {
      "entropy": 5.112243700027466,
      "epoch": 1.8263448045439359,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004665697531073489,
      "loss": 4.8603,
      "mean_token_accuracy": 0.2081417813897133,
      "num_tokens": 40233733.0,
      "step": 21865
    },
    {
      "entropy": 5.068102264404297,
      "epoch": 1.826762445706649,
      "grad_norm": 1.171875,
      "learning_rate": 0.000466554137593339,
      "loss": 4.8054,
      "mean_token_accuracy": 0.22109199017286302,
      "num_tokens": 40242897.0,
      "step": 21870
    },
    {
      "entropy": 5.1110978603363035,
      "epoch": 1.8271800868693617,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046653851872594334,
      "loss": 4.875,
      "mean_token_accuracy": 0.21630280762910842,
      "num_tokens": 40251936.0,
      "step": 21875
    },
    {
      "entropy": 5.1623961448669435,
      "epoch": 1.827597728032075,
      "grad_norm": 1.25,
      "learning_rate": 0.0004665228965054353,
      "loss": 4.9113,
      "mean_token_accuracy": 0.21317451894283296,
      "num_tokens": 40260295.0,
      "step": 21880
    },
    {
      "entropy": 5.152760410308838,
      "epoch": 1.8280153691947878,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004665072709320885,
      "loss": 4.8647,
      "mean_token_accuracy": 0.20946395099163057,
      "num_tokens": 40270058.0,
      "step": 21885
    },
    {
      "entropy": 5.064534902572632,
      "epoch": 1.8284330103575008,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046649164200617633,
      "loss": 4.7289,
      "mean_token_accuracy": 0.22806615084409715,
      "num_tokens": 40278599.0,
      "step": 21890
    },
    {
      "entropy": 5.1308451175689695,
      "epoch": 1.8288506515202139,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004664760097279725,
      "loss": 4.8701,
      "mean_token_accuracy": 0.20952766239643097,
      "num_tokens": 40288591.0,
      "step": 21895
    },
    {
      "entropy": 5.103038549423218,
      "epoch": 1.8292682926829267,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004664603740977506,
      "loss": 4.8506,
      "mean_token_accuracy": 0.21264761537313462,
      "num_tokens": 40297359.0,
      "step": 21900
    },
    {
      "entropy": 5.167620038986206,
      "epoch": 1.82968593384564,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046644473511578443,
      "loss": 4.9671,
      "mean_token_accuracy": 0.20356011390686035,
      "num_tokens": 40307130.0,
      "step": 21905
    },
    {
      "entropy": 5.122172164916992,
      "epoch": 1.8301035750083527,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046642909278234775,
      "loss": 4.7954,
      "mean_token_accuracy": 0.2207980841398239,
      "num_tokens": 40316065.0,
      "step": 21910
    },
    {
      "entropy": 5.100906896591186,
      "epoch": 1.8305212161710658,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004664134470977143,
      "loss": 4.8938,
      "mean_token_accuracy": 0.2051309049129486,
      "num_tokens": 40325808.0,
      "step": 21915
    },
    {
      "entropy": 5.139795017242432,
      "epoch": 1.8309388573337788,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046639779806215815,
      "loss": 4.8719,
      "mean_token_accuracy": 0.22043378800153732,
      "num_tokens": 40334632.0,
      "step": 21920
    },
    {
      "entropy": 5.113077878952026,
      "epoch": 1.8313564984964918,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004663821456759532,
      "loss": 4.889,
      "mean_token_accuracy": 0.20479063093662261,
      "num_tokens": 40343803.0,
      "step": 21925
    },
    {
      "entropy": 5.074895000457763,
      "epoch": 1.8317741396592049,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046636648993937326,
      "loss": 4.8271,
      "mean_token_accuracy": 0.22135789394378663,
      "num_tokens": 40352743.0,
      "step": 21930
    },
    {
      "entropy": 5.1538533687591555,
      "epoch": 1.8321917808219177,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004663508308526927,
      "loss": 4.9037,
      "mean_token_accuracy": 0.21067603379487992,
      "num_tokens": 40361421.0,
      "step": 21935
    },
    {
      "entropy": 5.142594289779663,
      "epoch": 1.832609421984631,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046633516841618547,
      "loss": 4.8329,
      "mean_token_accuracy": 0.21507835686206817,
      "num_tokens": 40369860.0,
      "step": 21940
    },
    {
      "entropy": 5.124264144897461,
      "epoch": 1.8330270631473438,
      "grad_norm": 1.25,
      "learning_rate": 0.0004663195026301258,
      "loss": 4.8867,
      "mean_token_accuracy": 0.21247074753046036,
      "num_tokens": 40379095.0,
      "step": 21945
    },
    {
      "entropy": 5.158783435821533,
      "epoch": 1.8334447043100568,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004663038334947879,
      "loss": 4.8756,
      "mean_token_accuracy": 0.21618425995111465,
      "num_tokens": 40388373.0,
      "step": 21950
    },
    {
      "entropy": 5.147570610046387,
      "epoch": 1.8338623454727698,
      "grad_norm": 1.25,
      "learning_rate": 0.0004662881610104461,
      "loss": 4.8792,
      "mean_token_accuracy": 0.21049902141094207,
      "num_tokens": 40398158.0,
      "step": 21955
    },
    {
      "entropy": 5.152856111526489,
      "epoch": 1.8342799866354826,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000466272485177375,
      "loss": 4.9707,
      "mean_token_accuracy": 0.20197321474552155,
      "num_tokens": 40407540.0,
      "step": 21960
    },
    {
      "entropy": 5.225804853439331,
      "epoch": 1.834697627798196,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046625680599584864,
      "loss": 5.0463,
      "mean_token_accuracy": 0.20278594493865967,
      "num_tokens": 40417445.0,
      "step": 21965
    },
    {
      "entropy": 5.2035231590271,
      "epoch": 1.8351152689609087,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046624112346614166,
      "loss": 4.9239,
      "mean_token_accuracy": 0.20612602382898332,
      "num_tokens": 40426873.0,
      "step": 21970
    },
    {
      "entropy": 5.169538164138794,
      "epoch": 1.8355329101236217,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004662254375885287,
      "loss": 4.8921,
      "mean_token_accuracy": 0.21256793439388275,
      "num_tokens": 40436072.0,
      "step": 21975
    },
    {
      "entropy": 5.103390884399414,
      "epoch": 1.8359505512863348,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046620974836328425,
      "loss": 4.8618,
      "mean_token_accuracy": 0.21358373612165452,
      "num_tokens": 40445599.0,
      "step": 21980
    },
    {
      "entropy": 5.133074617385864,
      "epoch": 1.8363681924490478,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004661940557906829,
      "loss": 4.9639,
      "mean_token_accuracy": 0.20970926582813262,
      "num_tokens": 40454982.0,
      "step": 21985
    },
    {
      "entropy": 5.0896830558776855,
      "epoch": 1.8367858336117608,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046617835987099957,
      "loss": 4.742,
      "mean_token_accuracy": 0.2244136080145836,
      "num_tokens": 40463794.0,
      "step": 21990
    },
    {
      "entropy": 5.070917224884033,
      "epoch": 1.8372034747744737,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000466162660604509,
      "loss": 4.8371,
      "mean_token_accuracy": 0.21821457147598267,
      "num_tokens": 40473139.0,
      "step": 21995
    },
    {
      "entropy": 5.093353509902954,
      "epoch": 1.837621115937187,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046614695799148575,
      "loss": 4.8626,
      "mean_token_accuracy": 0.2125195875763893,
      "num_tokens": 40481942.0,
      "step": 22000
    },
    {
      "entropy": 5.1703715324401855,
      "epoch": 1.8380387570998997,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046613125203220506,
      "loss": 4.8348,
      "mean_token_accuracy": 0.2165878966450691,
      "num_tokens": 40490428.0,
      "step": 22005
    },
    {
      "entropy": 5.222309875488281,
      "epoch": 1.8384563982626128,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046611554272694175,
      "loss": 4.9552,
      "mean_token_accuracy": 0.20605925619602203,
      "num_tokens": 40500185.0,
      "step": 22010
    },
    {
      "entropy": 5.059216403961182,
      "epoch": 1.8388740394253258,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004660998300759707,
      "loss": 4.8676,
      "mean_token_accuracy": 0.21523094177246094,
      "num_tokens": 40509205.0,
      "step": 22015
    },
    {
      "entropy": 5.075942516326904,
      "epoch": 1.8392916805880386,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046608411407956726,
      "loss": 4.8001,
      "mean_token_accuracy": 0.21264026015996934,
      "num_tokens": 40518910.0,
      "step": 22020
    },
    {
      "entropy": 5.137547445297241,
      "epoch": 1.8397093217507519,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046606839473800626,
      "loss": 4.886,
      "mean_token_accuracy": 0.21028826236724854,
      "num_tokens": 40526988.0,
      "step": 22025
    },
    {
      "entropy": 5.0663614749908445,
      "epoch": 1.8401269629134647,
      "grad_norm": 1.125,
      "learning_rate": 0.0004660526720515631,
      "loss": 4.811,
      "mean_token_accuracy": 0.22479122430086135,
      "num_tokens": 40536436.0,
      "step": 22030
    },
    {
      "entropy": 5.10438756942749,
      "epoch": 1.8405446040761777,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004660369460205128,
      "loss": 4.8822,
      "mean_token_accuracy": 0.2120947167277336,
      "num_tokens": 40546341.0,
      "step": 22035
    },
    {
      "entropy": 5.234272050857544,
      "epoch": 1.8409622452388907,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046602121664513104,
      "loss": 5.0033,
      "mean_token_accuracy": 0.20617773830890657,
      "num_tokens": 40556915.0,
      "step": 22040
    },
    {
      "entropy": 5.1554121494293215,
      "epoch": 1.8413798864016038,
      "grad_norm": 1.125,
      "learning_rate": 0.00046600548392569283,
      "loss": 4.8986,
      "mean_token_accuracy": 0.20847778618335724,
      "num_tokens": 40566881.0,
      "step": 22045
    },
    {
      "entropy": 5.016229295730591,
      "epoch": 1.8417975275643168,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046598974786247365,
      "loss": 4.7237,
      "mean_token_accuracy": 0.22007059007883073,
      "num_tokens": 40574845.0,
      "step": 22050
    },
    {
      "entropy": 5.043627071380615,
      "epoch": 1.8422151687270296,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004659740084557491,
      "loss": 4.77,
      "mean_token_accuracy": 0.22614704668521882,
      "num_tokens": 40583274.0,
      "step": 22055
    },
    {
      "entropy": 4.975346899032592,
      "epoch": 1.8426328098897429,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004659582657057947,
      "loss": 4.7624,
      "mean_token_accuracy": 0.21582767516374587,
      "num_tokens": 40592138.0,
      "step": 22060
    },
    {
      "entropy": 5.037513208389282,
      "epoch": 1.8430504510524557,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000465942519612886,
      "loss": 4.8241,
      "mean_token_accuracy": 0.21946845352649688,
      "num_tokens": 40600453.0,
      "step": 22065
    },
    {
      "entropy": 5.147690153121948,
      "epoch": 1.8434680922151687,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004659267701772986,
      "loss": 4.8907,
      "mean_token_accuracy": 0.20546946674585342,
      "num_tokens": 40610511.0,
      "step": 22070
    },
    {
      "entropy": 5.092477083206177,
      "epoch": 1.8438857333778818,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004659110173993083,
      "loss": 4.7853,
      "mean_token_accuracy": 0.22337360084056854,
      "num_tokens": 40619819.0,
      "step": 22075
    },
    {
      "entropy": 4.888289451599121,
      "epoch": 1.8443033745405946,
      "grad_norm": 1.25,
      "learning_rate": 0.0004658952612791908,
      "loss": 4.6547,
      "mean_token_accuracy": 0.22516819089651108,
      "num_tokens": 40628400.0,
      "step": 22080
    },
    {
      "entropy": 5.009007263183594,
      "epoch": 1.8447210157033078,
      "grad_norm": 1.203125,
      "learning_rate": 0.000465879501817222,
      "loss": 4.7951,
      "mean_token_accuracy": 0.21656694114208222,
      "num_tokens": 40637396.0,
      "step": 22085
    },
    {
      "entropy": 5.24489574432373,
      "epoch": 1.8451386568660206,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046586373901367774,
      "loss": 5.008,
      "mean_token_accuracy": 0.21020488142967225,
      "num_tokens": 40645902.0,
      "step": 22090
    },
    {
      "entropy": 5.231297349929809,
      "epoch": 1.8455562980287337,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000465847972868834,
      "loss": 4.9278,
      "mean_token_accuracy": 0.21246952563524246,
      "num_tokens": 40656382.0,
      "step": 22095
    },
    {
      "entropy": 5.142729091644287,
      "epoch": 1.8459739391914467,
      "grad_norm": 1.125,
      "learning_rate": 0.0004658322033829668,
      "loss": 4.9102,
      "mean_token_accuracy": 0.20588152706623078,
      "num_tokens": 40665471.0,
      "step": 22100
    },
    {
      "entropy": 5.063112688064575,
      "epoch": 1.8463915803541597,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046581643055635215,
      "loss": 4.8553,
      "mean_token_accuracy": 0.21545557230710982,
      "num_tokens": 40674668.0,
      "step": 22105
    },
    {
      "entropy": 5.100471544265747,
      "epoch": 1.8468092215168728,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046580065438926624,
      "loss": 4.7995,
      "mean_token_accuracy": 0.2151978313922882,
      "num_tokens": 40683779.0,
      "step": 22110
    },
    {
      "entropy": 5.116083812713623,
      "epoch": 1.8472268626795856,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004657848748819851,
      "loss": 4.8907,
      "mean_token_accuracy": 0.2153159961104393,
      "num_tokens": 40692412.0,
      "step": 22115
    },
    {
      "entropy": 5.025872611999512,
      "epoch": 1.8476445038422988,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046576909203478517,
      "loss": 4.7114,
      "mean_token_accuracy": 0.22770529091358185,
      "num_tokens": 40701652.0,
      "step": 22120
    },
    {
      "entropy": 5.245446348190308,
      "epoch": 1.8480621450050116,
      "grad_norm": 1.265625,
      "learning_rate": 0.00046575330584794263,
      "loss": 5.0533,
      "mean_token_accuracy": 0.20291892141103746,
      "num_tokens": 40711149.0,
      "step": 22125
    },
    {
      "entropy": 5.199782419204712,
      "epoch": 1.8484797861677247,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004657375163217338,
      "loss": 4.8809,
      "mean_token_accuracy": 0.21547014266252518,
      "num_tokens": 40721014.0,
      "step": 22130
    },
    {
      "entropy": 5.088195514678955,
      "epoch": 1.8488974273304377,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046572172345643527,
      "loss": 4.8593,
      "mean_token_accuracy": 0.21424803286790847,
      "num_tokens": 40730645.0,
      "step": 22135
    },
    {
      "entropy": 5.025973701477051,
      "epoch": 1.8493150684931505,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046570592725232333,
      "loss": 4.7599,
      "mean_token_accuracy": 0.22212646901607513,
      "num_tokens": 40739620.0,
      "step": 22140
    },
    {
      "entropy": 5.0287047863006595,
      "epoch": 1.8497327096558638,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004656901277096746,
      "loss": 4.8286,
      "mean_token_accuracy": 0.2218078076839447,
      "num_tokens": 40748840.0,
      "step": 22145
    },
    {
      "entropy": 5.212420797348022,
      "epoch": 1.8501503508185766,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004656743248287656,
      "loss": 4.9958,
      "mean_token_accuracy": 0.20672837495803834,
      "num_tokens": 40757798.0,
      "step": 22150
    },
    {
      "entropy": 5.068237972259522,
      "epoch": 1.8505679919812896,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004656585186098731,
      "loss": 4.8035,
      "mean_token_accuracy": 0.2186818391084671,
      "num_tokens": 40766853.0,
      "step": 22155
    },
    {
      "entropy": 5.046643352508545,
      "epoch": 1.8509856331440027,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004656427090532737,
      "loss": 4.799,
      "mean_token_accuracy": 0.22340796440839766,
      "num_tokens": 40775232.0,
      "step": 22160
    },
    {
      "entropy": 5.17793083190918,
      "epoch": 1.8514032743067157,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046562689615924427,
      "loss": 4.8581,
      "mean_token_accuracy": 0.21663658022880555,
      "num_tokens": 40783759.0,
      "step": 22165
    },
    {
      "entropy": 5.180886268615723,
      "epoch": 1.8518209154694287,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004656110799280615,
      "loss": 4.8996,
      "mean_token_accuracy": 0.20822956562042236,
      "num_tokens": 40792819.0,
      "step": 22170
    },
    {
      "entropy": 5.103601789474487,
      "epoch": 1.8522385566321415,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046559526036000235,
      "loss": 4.7971,
      "mean_token_accuracy": 0.2158699169754982,
      "num_tokens": 40801956.0,
      "step": 22175
    },
    {
      "entropy": 5.094752931594849,
      "epoch": 1.8526561977948548,
      "grad_norm": 1.28125,
      "learning_rate": 0.00046557943745534374,
      "loss": 4.8792,
      "mean_token_accuracy": 0.21206091195344925,
      "num_tokens": 40812439.0,
      "step": 22180
    },
    {
      "entropy": 5.164037561416626,
      "epoch": 1.8530738389575676,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046556361121436265,
      "loss": 4.9533,
      "mean_token_accuracy": 0.20713705867528914,
      "num_tokens": 40822712.0,
      "step": 22185
    },
    {
      "entropy": 5.129747438430786,
      "epoch": 1.8534914801202806,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046554778163733626,
      "loss": 4.8288,
      "mean_token_accuracy": 0.21348964869976045,
      "num_tokens": 40831242.0,
      "step": 22190
    },
    {
      "entropy": 5.03889274597168,
      "epoch": 1.8539091212829937,
      "grad_norm": 1.28125,
      "learning_rate": 0.00046553194872454146,
      "loss": 4.7652,
      "mean_token_accuracy": 0.22210224270820617,
      "num_tokens": 40839754.0,
      "step": 22195
    },
    {
      "entropy": 5.134660720825195,
      "epoch": 1.8543267624457065,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046551611247625565,
      "loss": 4.9244,
      "mean_token_accuracy": 0.21235694885253906,
      "num_tokens": 40848925.0,
      "step": 22200
    },
    {
      "entropy": 5.10527982711792,
      "epoch": 1.8547444036084197,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004655002728927559,
      "loss": 4.8515,
      "mean_token_accuracy": 0.21300016343593597,
      "num_tokens": 40857784.0,
      "step": 22205
    },
    {
      "entropy": 4.987250661849975,
      "epoch": 1.8551620447711326,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046548442997431966,
      "loss": 4.681,
      "mean_token_accuracy": 0.2309543952345848,
      "num_tokens": 40866607.0,
      "step": 22210
    },
    {
      "entropy": 5.218245506286621,
      "epoch": 1.8555796859338456,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004654685837212241,
      "loss": 4.8917,
      "mean_token_accuracy": 0.21215408444404601,
      "num_tokens": 40876730.0,
      "step": 22215
    },
    {
      "entropy": 5.084536790847778,
      "epoch": 1.8559973270965586,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004654527341337468,
      "loss": 4.8471,
      "mean_token_accuracy": 0.20268765687942505,
      "num_tokens": 40886482.0,
      "step": 22220
    },
    {
      "entropy": 5.134303951263428,
      "epoch": 1.8564149682592717,
      "grad_norm": 1.125,
      "learning_rate": 0.00046543688121216504,
      "loss": 4.9397,
      "mean_token_accuracy": 0.20707745999097824,
      "num_tokens": 40896704.0,
      "step": 22225
    },
    {
      "entropy": 5.113797378540039,
      "epoch": 1.8568326094219847,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004654210249567565,
      "loss": 4.8461,
      "mean_token_accuracy": 0.21142319440841675,
      "num_tokens": 40906665.0,
      "step": 22230
    },
    {
      "entropy": 5.1465380668640135,
      "epoch": 1.8572502505846975,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004654051653677988,
      "loss": 4.8863,
      "mean_token_accuracy": 0.21107357442379,
      "num_tokens": 40915838.0,
      "step": 22235
    },
    {
      "entropy": 5.0709034442901615,
      "epoch": 1.8576678917474108,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046538930244556927,
      "loss": 4.8236,
      "mean_token_accuracy": 0.20971937775611876,
      "num_tokens": 40924571.0,
      "step": 22240
    },
    {
      "entropy": 5.087313318252564,
      "epoch": 1.8580855329101236,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046537343619034593,
      "loss": 4.8211,
      "mean_token_accuracy": 0.21407225877046585,
      "num_tokens": 40934421.0,
      "step": 22245
    },
    {
      "entropy": 5.052290630340576,
      "epoch": 1.8585031740728366,
      "grad_norm": 1.421875,
      "learning_rate": 0.00046535756660240644,
      "loss": 4.6834,
      "mean_token_accuracy": 0.2259597510099411,
      "num_tokens": 40943090.0,
      "step": 22250
    },
    {
      "entropy": 5.055368232727051,
      "epoch": 1.8589208152355496,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004653416936820287,
      "loss": 4.8766,
      "mean_token_accuracy": 0.21371446400880814,
      "num_tokens": 40952568.0,
      "step": 22255
    },
    {
      "entropy": 5.0258016109466555,
      "epoch": 1.8593384563982625,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004653258174294904,
      "loss": 4.8458,
      "mean_token_accuracy": 0.21599810868501662,
      "num_tokens": 40961122.0,
      "step": 22260
    },
    {
      "entropy": 5.153583431243897,
      "epoch": 1.8597560975609757,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004653099378450696,
      "loss": 4.8918,
      "mean_token_accuracy": 0.2027151107788086,
      "num_tokens": 40971117.0,
      "step": 22265
    },
    {
      "entropy": 5.135557985305786,
      "epoch": 1.8601737387236885,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046529405492904427,
      "loss": 4.8638,
      "mean_token_accuracy": 0.2120377704501152,
      "num_tokens": 40981151.0,
      "step": 22270
    },
    {
      "entropy": 5.025947713851929,
      "epoch": 1.8605913798864016,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004652781686816925,
      "loss": 4.7679,
      "mean_token_accuracy": 0.21749859303236008,
      "num_tokens": 40990481.0,
      "step": 22275
    },
    {
      "entropy": 5.039278030395508,
      "epoch": 1.8610090210491146,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004652622791032922,
      "loss": 4.7933,
      "mean_token_accuracy": 0.2136596620082855,
      "num_tokens": 40999360.0,
      "step": 22280
    },
    {
      "entropy": 5.133011102676392,
      "epoch": 1.8614266622118276,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046524638619412187,
      "loss": 4.9326,
      "mean_token_accuracy": 0.20364575684070588,
      "num_tokens": 41009342.0,
      "step": 22285
    },
    {
      "entropy": 5.180753135681153,
      "epoch": 1.8618443033745407,
      "grad_norm": 1.125,
      "learning_rate": 0.00046523048995445945,
      "loss": 4.8172,
      "mean_token_accuracy": 0.21342293173074722,
      "num_tokens": 41018219.0,
      "step": 22290
    },
    {
      "entropy": 5.117642498016357,
      "epoch": 1.8622619445372535,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046521459038458327,
      "loss": 4.8835,
      "mean_token_accuracy": 0.21393003463745117,
      "num_tokens": 41027339.0,
      "step": 22295
    },
    {
      "entropy": 5.151702499389648,
      "epoch": 1.8626795856999667,
      "grad_norm": 1.125,
      "learning_rate": 0.00046519868748477186,
      "loss": 4.887,
      "mean_token_accuracy": 0.2147297665476799,
      "num_tokens": 41036748.0,
      "step": 22300
    },
    {
      "entropy": 5.084293985366822,
      "epoch": 1.8630972268626795,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046518278125530336,
      "loss": 4.8451,
      "mean_token_accuracy": 0.21918318569660186,
      "num_tokens": 41046064.0,
      "step": 22305
    },
    {
      "entropy": 5.058805465698242,
      "epoch": 1.8635148680253926,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004651668716964564,
      "loss": 4.8559,
      "mean_token_accuracy": 0.2114357218146324,
      "num_tokens": 41055784.0,
      "step": 22310
    },
    {
      "entropy": 5.140916442871093,
      "epoch": 1.8639325091881056,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046515095880850937,
      "loss": 4.8098,
      "mean_token_accuracy": 0.21699164509773256,
      "num_tokens": 41065585.0,
      "step": 22315
    },
    {
      "entropy": 5.012788963317871,
      "epoch": 1.8643501503508184,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046513504259174104,
      "loss": 4.7689,
      "mean_token_accuracy": 0.22625656872987748,
      "num_tokens": 41073531.0,
      "step": 22320
    },
    {
      "entropy": 5.100238180160522,
      "epoch": 1.8647677915135317,
      "grad_norm": 1.125,
      "learning_rate": 0.0004651191230464299,
      "loss": 4.8972,
      "mean_token_accuracy": 0.21006526201963424,
      "num_tokens": 41082370.0,
      "step": 22325
    },
    {
      "entropy": 5.024028396606445,
      "epoch": 1.8651854326762445,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004651032001728546,
      "loss": 4.832,
      "mean_token_accuracy": 0.2101888909935951,
      "num_tokens": 41091025.0,
      "step": 22330
    },
    {
      "entropy": 5.161368322372437,
      "epoch": 1.8656030738389575,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004650872739712941,
      "loss": 4.9333,
      "mean_token_accuracy": 0.21031830608844757,
      "num_tokens": 41100846.0,
      "step": 22335
    },
    {
      "entropy": 5.178827524185181,
      "epoch": 1.8660207150016705,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046507134444202694,
      "loss": 4.7935,
      "mean_token_accuracy": 0.2199384406208992,
      "num_tokens": 41109885.0,
      "step": 22340
    },
    {
      "entropy": 5.055418062210083,
      "epoch": 1.8664383561643836,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004650554115853321,
      "loss": 4.8238,
      "mean_token_accuracy": 0.21648596674203874,
      "num_tokens": 41118478.0,
      "step": 22345
    },
    {
      "entropy": 5.028544807434082,
      "epoch": 1.8668559973270966,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046503947540148857,
      "loss": 4.7621,
      "mean_token_accuracy": 0.2238980382680893,
      "num_tokens": 41127605.0,
      "step": 22350
    },
    {
      "entropy": 5.127436923980713,
      "epoch": 1.8672736384898094,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046502353589077523,
      "loss": 4.9023,
      "mean_token_accuracy": 0.20747800171375275,
      "num_tokens": 41137850.0,
      "step": 22355
    },
    {
      "entropy": 5.099642276763916,
      "epoch": 1.8676912796525227,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046500759305347116,
      "loss": 4.856,
      "mean_token_accuracy": 0.22127819657325745,
      "num_tokens": 41147452.0,
      "step": 22360
    },
    {
      "entropy": 5.040412950515747,
      "epoch": 1.8681089208152355,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004649916468898555,
      "loss": 4.7378,
      "mean_token_accuracy": 0.2167290598154068,
      "num_tokens": 41155946.0,
      "step": 22365
    },
    {
      "entropy": 5.113527441024781,
      "epoch": 1.8685265619779485,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004649756974002073,
      "loss": 4.9878,
      "mean_token_accuracy": 0.20773232132196426,
      "num_tokens": 41165048.0,
      "step": 22370
    },
    {
      "entropy": 5.035595655441284,
      "epoch": 1.8689442031406616,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004649597445848059,
      "loss": 4.6777,
      "mean_token_accuracy": 0.2288101702928543,
      "num_tokens": 41174026.0,
      "step": 22375
    },
    {
      "entropy": 5.143392848968506,
      "epoch": 1.8693618443033744,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004649437884439305,
      "loss": 4.9015,
      "mean_token_accuracy": 0.2155812382698059,
      "num_tokens": 41183190.0,
      "step": 22380
    },
    {
      "entropy": 4.986326885223389,
      "epoch": 1.8697794854660876,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004649278289778604,
      "loss": 4.7954,
      "mean_token_accuracy": 0.21391381025314332,
      "num_tokens": 41192070.0,
      "step": 22385
    },
    {
      "entropy": 5.004288673400879,
      "epoch": 1.8701971266288004,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046491186618687514,
      "loss": 4.7579,
      "mean_token_accuracy": 0.21944290697574614,
      "num_tokens": 41201166.0,
      "step": 22390
    },
    {
      "entropy": 5.113340997695923,
      "epoch": 1.8706147677915135,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046489590007125397,
      "loss": 4.8343,
      "mean_token_accuracy": 0.2201665997505188,
      "num_tokens": 41208998.0,
      "step": 22395
    },
    {
      "entropy": 5.126497030258179,
      "epoch": 1.8710324089542265,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004648799306312766,
      "loss": 4.7768,
      "mean_token_accuracy": 0.22365565299987794,
      "num_tokens": 41218206.0,
      "step": 22400
    },
    {
      "entropy": 5.018629884719848,
      "epoch": 1.8714500501169395,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046486395786722237,
      "loss": 4.7953,
      "mean_token_accuracy": 0.21309429109096528,
      "num_tokens": 41226345.0,
      "step": 22405
    },
    {
      "entropy": 5.075983953475952,
      "epoch": 1.8718676912796526,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004648479817793711,
      "loss": 4.8614,
      "mean_token_accuracy": 0.2139805406332016,
      "num_tokens": 41235641.0,
      "step": 22410
    },
    {
      "entropy": 5.122745752334595,
      "epoch": 1.8722853324423654,
      "grad_norm": 1.125,
      "learning_rate": 0.0004648320023680023,
      "loss": 4.8563,
      "mean_token_accuracy": 0.21150009334087372,
      "num_tokens": 41245308.0,
      "step": 22415
    },
    {
      "entropy": 5.109957075119018,
      "epoch": 1.8727029736050786,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046481601963339593,
      "loss": 4.869,
      "mean_token_accuracy": 0.21773480325937272,
      "num_tokens": 41254030.0,
      "step": 22420
    },
    {
      "entropy": 5.118829917907715,
      "epoch": 1.8731206147677915,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046480003357583154,
      "loss": 4.8697,
      "mean_token_accuracy": 0.2143391862511635,
      "num_tokens": 41263449.0,
      "step": 22425
    },
    {
      "entropy": 5.204363918304443,
      "epoch": 1.8735382559305045,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004647840441955892,
      "loss": 4.9283,
      "mean_token_accuracy": 0.21015281826257706,
      "num_tokens": 41272379.0,
      "step": 22430
    },
    {
      "entropy": 5.0765594959259035,
      "epoch": 1.8739558970932175,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004647680514929486,
      "loss": 4.8762,
      "mean_token_accuracy": 0.2178833931684494,
      "num_tokens": 41281384.0,
      "step": 22435
    },
    {
      "entropy": 5.07415738105774,
      "epoch": 1.8743735382559303,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046475205546818996,
      "loss": 4.7653,
      "mean_token_accuracy": 0.22708216905593873,
      "num_tokens": 41290662.0,
      "step": 22440
    },
    {
      "entropy": 5.0838111400604244,
      "epoch": 1.8747911794186436,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004647360561215931,
      "loss": 4.8603,
      "mean_token_accuracy": 0.2124269887804985,
      "num_tokens": 41300278.0,
      "step": 22445
    },
    {
      "entropy": 5.147473669052124,
      "epoch": 1.8752088205813564,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046472005345343814,
      "loss": 4.9271,
      "mean_token_accuracy": 0.20771690160036088,
      "num_tokens": 41309456.0,
      "step": 22450
    },
    {
      "entropy": 5.147909021377563,
      "epoch": 1.8756264617440697,
      "grad_norm": 1.125,
      "learning_rate": 0.0004647040474640054,
      "loss": 4.9351,
      "mean_token_accuracy": 0.20801565051078796,
      "num_tokens": 41318550.0,
      "step": 22455
    },
    {
      "entropy": 5.124496650695801,
      "epoch": 1.8760441029067825,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004646880381535749,
      "loss": 4.9151,
      "mean_token_accuracy": 0.2140699580311775,
      "num_tokens": 41327867.0,
      "step": 22460
    },
    {
      "entropy": 5.040001296997071,
      "epoch": 1.8764617440694955,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000464672025522427,
      "loss": 4.7269,
      "mean_token_accuracy": 0.22604029029607772,
      "num_tokens": 41337151.0,
      "step": 22465
    },
    {
      "entropy": 4.997120571136475,
      "epoch": 1.8768793852322085,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000464656009570842,
      "loss": 4.7311,
      "mean_token_accuracy": 0.22255251854658126,
      "num_tokens": 41346205.0,
      "step": 22470
    },
    {
      "entropy": 5.117643880844116,
      "epoch": 1.8772970263949214,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004646399902991002,
      "loss": 4.8292,
      "mean_token_accuracy": 0.21735401153564454,
      "num_tokens": 41354288.0,
      "step": 22475
    },
    {
      "entropy": 5.117129516601563,
      "epoch": 1.8777146675576346,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004646239677074821,
      "loss": 4.829,
      "mean_token_accuracy": 0.2135946422815323,
      "num_tokens": 41363366.0,
      "step": 22480
    },
    {
      "entropy": 5.059537267684936,
      "epoch": 1.8781323087203474,
      "grad_norm": 1.25,
      "learning_rate": 0.00046460794179626827,
      "loss": 4.8229,
      "mean_token_accuracy": 0.21922830045223235,
      "num_tokens": 41371778.0,
      "step": 22485
    },
    {
      "entropy": 5.056646537780762,
      "epoch": 1.8785499498830605,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004645919125657391,
      "loss": 4.7426,
      "mean_token_accuracy": 0.22941609472036362,
      "num_tokens": 41380035.0,
      "step": 22490
    },
    {
      "entropy": 5.190626430511474,
      "epoch": 1.8789675910457735,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004645758800161754,
      "loss": 4.981,
      "mean_token_accuracy": 0.20044645816087722,
      "num_tokens": 41389819.0,
      "step": 22495
    },
    {
      "entropy": 5.121442890167236,
      "epoch": 1.8793852322084865,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004645598441478577,
      "loss": 4.9081,
      "mean_token_accuracy": 0.20799740701913833,
      "num_tokens": 41398704.0,
      "step": 22500
    },
    {
      "entropy": 5.1425315856933596,
      "epoch": 1.8798028733711996,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046454380496106665,
      "loss": 4.8825,
      "mean_token_accuracy": 0.21146571338176728,
      "num_tokens": 41406634.0,
      "step": 22505
    },
    {
      "entropy": 5.093302536010742,
      "epoch": 1.8802205145339124,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004645277624560832,
      "loss": 4.8868,
      "mean_token_accuracy": 0.20376602113246917,
      "num_tokens": 41416164.0,
      "step": 22510
    },
    {
      "entropy": 5.131656217575073,
      "epoch": 1.8806381556966256,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046451171663318815,
      "loss": 4.9168,
      "mean_token_accuracy": 0.20708692967891693,
      "num_tokens": 41426100.0,
      "step": 22515
    },
    {
      "entropy": 5.250997257232666,
      "epoch": 1.8810557968593384,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046449566749266237,
      "loss": 5.0152,
      "mean_token_accuracy": 0.2011991873383522,
      "num_tokens": 41435901.0,
      "step": 22520
    },
    {
      "entropy": 5.134066152572632,
      "epoch": 1.8814734380220515,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004644796150347868,
      "loss": 4.8502,
      "mean_token_accuracy": 0.21899795830249785,
      "num_tokens": 41444416.0,
      "step": 22525
    },
    {
      "entropy": 5.070746326446534,
      "epoch": 1.8818910791847645,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046446355925984257,
      "loss": 4.8786,
      "mean_token_accuracy": 0.2182386562228203,
      "num_tokens": 41454477.0,
      "step": 22530
    },
    {
      "entropy": 5.151042652130127,
      "epoch": 1.8823087203474773,
      "grad_norm": 1.125,
      "learning_rate": 0.00046444750016811064,
      "loss": 4.9361,
      "mean_token_accuracy": 0.2073918804526329,
      "num_tokens": 41464156.0,
      "step": 22535
    },
    {
      "entropy": 5.17722978591919,
      "epoch": 1.8827263615101906,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004644314377598722,
      "loss": 4.8647,
      "mean_token_accuracy": 0.21391750872135162,
      "num_tokens": 41473236.0,
      "step": 22540
    },
    {
      "entropy": 5.098812007904053,
      "epoch": 1.8831440026729034,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004644153720354084,
      "loss": 4.8457,
      "mean_token_accuracy": 0.2125965803861618,
      "num_tokens": 41482562.0,
      "step": 22545
    },
    {
      "entropy": 5.14416823387146,
      "epoch": 1.8835616438356164,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004643993029950005,
      "loss": 4.8455,
      "mean_token_accuracy": 0.21040482670068741,
      "num_tokens": 41490405.0,
      "step": 22550
    },
    {
      "entropy": 5.183311176300049,
      "epoch": 1.8839792849983295,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046438323063892985,
      "loss": 4.9588,
      "mean_token_accuracy": 0.20378762632608413,
      "num_tokens": 41500188.0,
      "step": 22555
    },
    {
      "entropy": 5.121012353897095,
      "epoch": 1.8843969261610425,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046436715496747774,
      "loss": 4.899,
      "mean_token_accuracy": 0.204659204185009,
      "num_tokens": 41509736.0,
      "step": 22560
    },
    {
      "entropy": 5.126393461227417,
      "epoch": 1.8848145673237555,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004643510759809257,
      "loss": 4.8692,
      "mean_token_accuracy": 0.20560440868139268,
      "num_tokens": 41519795.0,
      "step": 22565
    },
    {
      "entropy": 5.202544116973877,
      "epoch": 1.8852322084864683,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004643349936795551,
      "loss": 4.8745,
      "mean_token_accuracy": 0.20864115357398988,
      "num_tokens": 41529602.0,
      "step": 22570
    },
    {
      "entropy": 5.044680213928222,
      "epoch": 1.8856498496491816,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004643189080636475,
      "loss": 4.8845,
      "mean_token_accuracy": 0.21076679676771165,
      "num_tokens": 41538850.0,
      "step": 22575
    },
    {
      "entropy": 5.051567220687867,
      "epoch": 1.8860674908118944,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046430281913348466,
      "loss": 4.8082,
      "mean_token_accuracy": 0.21603965908288955,
      "num_tokens": 41548099.0,
      "step": 22580
    },
    {
      "entropy": 5.228936052322387,
      "epoch": 1.8864851319746074,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004642867268893481,
      "loss": 5.0209,
      "mean_token_accuracy": 0.2057591572403908,
      "num_tokens": 41557288.0,
      "step": 22585
    },
    {
      "entropy": 5.05746693611145,
      "epoch": 1.8869027731373205,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004642706313315194,
      "loss": 4.7723,
      "mean_token_accuracy": 0.23004979640245438,
      "num_tokens": 41565938.0,
      "step": 22590
    },
    {
      "entropy": 5.111138200759887,
      "epoch": 1.8873204143000333,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046425453246028056,
      "loss": 4.912,
      "mean_token_accuracy": 0.20958805382251738,
      "num_tokens": 41574984.0,
      "step": 22595
    },
    {
      "entropy": 5.1245002269744875,
      "epoch": 1.8877380554627465,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004642384302759133,
      "loss": 4.7905,
      "mean_token_accuracy": 0.21973819732666017,
      "num_tokens": 41584020.0,
      "step": 22600
    },
    {
      "entropy": 5.151034641265869,
      "epoch": 1.8881556966254593,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046422232477869957,
      "loss": 4.8969,
      "mean_token_accuracy": 0.21075182557106018,
      "num_tokens": 41594146.0,
      "step": 22605
    },
    {
      "entropy": 5.135475111007691,
      "epoch": 1.8885733377881724,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004642062159689213,
      "loss": 4.9278,
      "mean_token_accuracy": 0.2120328813791275,
      "num_tokens": 41603469.0,
      "step": 22610
    },
    {
      "entropy": 5.132319641113281,
      "epoch": 1.8889909789508854,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046419010384686043,
      "loss": 4.8842,
      "mean_token_accuracy": 0.2133372575044632,
      "num_tokens": 41613940.0,
      "step": 22615
    },
    {
      "entropy": 5.137695693969727,
      "epoch": 1.8894086201135984,
      "grad_norm": 1.21875,
      "learning_rate": 0.000464173988412799,
      "loss": 4.8345,
      "mean_token_accuracy": 0.21589296758174897,
      "num_tokens": 41623153.0,
      "step": 22620
    },
    {
      "entropy": 5.067892837524414,
      "epoch": 1.8898262612763115,
      "grad_norm": 1.328125,
      "learning_rate": 0.00046415786966701926,
      "loss": 4.8683,
      "mean_token_accuracy": 0.21179263442754745,
      "num_tokens": 41632396.0,
      "step": 22625
    },
    {
      "entropy": 5.021264219284058,
      "epoch": 1.8902439024390243,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004641417476098034,
      "loss": 4.7539,
      "mean_token_accuracy": 0.22307133823633193,
      "num_tokens": 41640727.0,
      "step": 22630
    },
    {
      "entropy": 4.927138042449951,
      "epoch": 1.8906615436017375,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004641256222414335,
      "loss": 4.644,
      "mean_token_accuracy": 0.22577416747808457,
      "num_tokens": 41648805.0,
      "step": 22635
    },
    {
      "entropy": 5.103774976730347,
      "epoch": 1.8910791847644504,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004641094935621919,
      "loss": 4.8748,
      "mean_token_accuracy": 0.21979361027479172,
      "num_tokens": 41657356.0,
      "step": 22640
    },
    {
      "entropy": 5.160951042175293,
      "epoch": 1.8914968259271634,
      "grad_norm": 1.203125,
      "learning_rate": 0.000464093361572361,
      "loss": 4.8502,
      "mean_token_accuracy": 0.21155091226100922,
      "num_tokens": 41666525.0,
      "step": 22645
    },
    {
      "entropy": 5.02749719619751,
      "epoch": 1.8919144670898764,
      "grad_norm": 1.265625,
      "learning_rate": 0.00046407722627222326,
      "loss": 4.7964,
      "mean_token_accuracy": 0.21814396977424622,
      "num_tokens": 41675113.0,
      "step": 22650
    },
    {
      "entropy": 5.170849847793579,
      "epoch": 1.8923321082525892,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046406108766206103,
      "loss": 4.9503,
      "mean_token_accuracy": 0.20550136566162108,
      "num_tokens": 41684283.0,
      "step": 22655
    },
    {
      "entropy": 5.105032873153687,
      "epoch": 1.8927497494153025,
      "grad_norm": 0.984375,
      "learning_rate": 0.00046404494574215683,
      "loss": 4.7979,
      "mean_token_accuracy": 0.2213537573814392,
      "num_tokens": 41693751.0,
      "step": 22660
    },
    {
      "entropy": 5.058942413330078,
      "epoch": 1.8931673905780153,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004640288005127933,
      "loss": 4.8082,
      "mean_token_accuracy": 0.2195175588130951,
      "num_tokens": 41702776.0,
      "step": 22665
    },
    {
      "entropy": 5.034315824508667,
      "epoch": 1.8935850317407283,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004640126519742532,
      "loss": 4.8073,
      "mean_token_accuracy": 0.21482089161872864,
      "num_tokens": 41711567.0,
      "step": 22670
    },
    {
      "entropy": 5.140657043457031,
      "epoch": 1.8940026729034414,
      "grad_norm": 1.46875,
      "learning_rate": 0.000463996500126819,
      "loss": 4.8667,
      "mean_token_accuracy": 0.2196657806634903,
      "num_tokens": 41720784.0,
      "step": 22675
    },
    {
      "entropy": 5.065290403366089,
      "epoch": 1.8944203140661544,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004639803449707737,
      "loss": 4.7796,
      "mean_token_accuracy": 0.21480634808540344,
      "num_tokens": 41730775.0,
      "step": 22680
    },
    {
      "entropy": 5.04120626449585,
      "epoch": 1.8948379552288674,
      "grad_norm": 1.25,
      "learning_rate": 0.00046396418650639993,
      "loss": 4.7846,
      "mean_token_accuracy": 0.22019775360822677,
      "num_tokens": 41740151.0,
      "step": 22685
    },
    {
      "entropy": 5.121014881134033,
      "epoch": 1.8952555963915803,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046394802473398065,
      "loss": 4.8933,
      "mean_token_accuracy": 0.2159357413649559,
      "num_tokens": 41748469.0,
      "step": 22690
    },
    {
      "entropy": 5.154291105270386,
      "epoch": 1.8956732375542935,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004639318596537986,
      "loss": 4.8552,
      "mean_token_accuracy": 0.20650867074728013,
      "num_tokens": 41757771.0,
      "step": 22695
    },
    {
      "entropy": 5.03644552230835,
      "epoch": 1.8960908787170063,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000463915691266137,
      "loss": 4.803,
      "mean_token_accuracy": 0.21359065622091294,
      "num_tokens": 41766979.0,
      "step": 22700
    },
    {
      "entropy": 4.995301818847656,
      "epoch": 1.8965085198797194,
      "grad_norm": 1.25,
      "learning_rate": 0.0004638995195712789,
      "loss": 4.7546,
      "mean_token_accuracy": 0.22130293846130372,
      "num_tokens": 41775094.0,
      "step": 22705
    },
    {
      "entropy": 5.188891172409058,
      "epoch": 1.8969261610424324,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004638833445695073,
      "loss": 4.9245,
      "mean_token_accuracy": 0.21546248644590377,
      "num_tokens": 41784608.0,
      "step": 22710
    },
    {
      "entropy": 5.139255428314209,
      "epoch": 1.8973438022051452,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046386716626110544,
      "loss": 4.928,
      "mean_token_accuracy": 0.2132851853966713,
      "num_tokens": 41793789.0,
      "step": 22715
    },
    {
      "entropy": 5.113216066360474,
      "epoch": 1.8977614433678585,
      "grad_norm": 1.296875,
      "learning_rate": 0.00046385098464635645,
      "loss": 4.8859,
      "mean_token_accuracy": 0.20858197957277297,
      "num_tokens": 41802928.0,
      "step": 22720
    },
    {
      "entropy": 5.205642557144165,
      "epoch": 1.8981790845305713,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004638347997255436,
      "loss": 4.9831,
      "mean_token_accuracy": 0.2045236721634865,
      "num_tokens": 41812231.0,
      "step": 22725
    },
    {
      "entropy": 5.128707504272461,
      "epoch": 1.8985967256932843,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004638186114989504,
      "loss": 4.8819,
      "mean_token_accuracy": 0.21722643822431564,
      "num_tokens": 41820665.0,
      "step": 22730
    },
    {
      "entropy": 5.123510217666626,
      "epoch": 1.8990143668559973,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004638024199668601,
      "loss": 4.8318,
      "mean_token_accuracy": 0.2134073093533516,
      "num_tokens": 41829568.0,
      "step": 22735
    },
    {
      "entropy": 5.051218843460083,
      "epoch": 1.8994320080187104,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004637862251295561,
      "loss": 4.7567,
      "mean_token_accuracy": 0.2146673947572708,
      "num_tokens": 41838360.0,
      "step": 22740
    },
    {
      "entropy": 5.063341093063355,
      "epoch": 1.8998496491814234,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000463770026987322,
      "loss": 4.7733,
      "mean_token_accuracy": 0.21443051248788833,
      "num_tokens": 41847050.0,
      "step": 22745
    },
    {
      "entropy": 5.139056539535522,
      "epoch": 1.9002672903441362,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00046375382554044136,
      "loss": 4.8688,
      "mean_token_accuracy": 0.21512419134378433,
      "num_tokens": 41856379.0,
      "step": 22750
    },
    {
      "entropy": 5.106476354598999,
      "epoch": 1.9006849315068495,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004637376207891978,
      "loss": 4.8005,
      "mean_token_accuracy": 0.22258548438549042,
      "num_tokens": 41865139.0,
      "step": 22755
    },
    {
      "entropy": 5.077171516418457,
      "epoch": 1.9011025726695623,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000463721412733875,
      "loss": 4.8445,
      "mean_token_accuracy": 0.221068774163723,
      "num_tokens": 41873955.0,
      "step": 22760
    },
    {
      "entropy": 5.096725940704346,
      "epoch": 1.9015202138322753,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046370520137475673,
      "loss": 4.8936,
      "mean_token_accuracy": 0.2121592253446579,
      "num_tokens": 41883771.0,
      "step": 22765
    },
    {
      "entropy": 5.144817543029785,
      "epoch": 1.9019378549949884,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004636889867121268,
      "loss": 4.8603,
      "mean_token_accuracy": 0.21368387937545777,
      "num_tokens": 41893411.0,
      "step": 22770
    },
    {
      "entropy": 5.037040424346924,
      "epoch": 1.9023554961577012,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046367276874626894,
      "loss": 4.8278,
      "mean_token_accuracy": 0.21544245779514312,
      "num_tokens": 41903612.0,
      "step": 22775
    },
    {
      "entropy": 5.112432289123535,
      "epoch": 1.9027731373204144,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004636565474774673,
      "loss": 4.8129,
      "mean_token_accuracy": 0.22226409763097763,
      "num_tokens": 41912754.0,
      "step": 22780
    },
    {
      "entropy": 5.129266262054443,
      "epoch": 1.9031907784831272,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046364032290600556,
      "loss": 4.8455,
      "mean_token_accuracy": 0.21029406040906906,
      "num_tokens": 41922854.0,
      "step": 22785
    },
    {
      "entropy": 5.075244569778443,
      "epoch": 1.9036084196458403,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000463624095032168,
      "loss": 4.8782,
      "mean_token_accuracy": 0.21639678925275802,
      "num_tokens": 41932259.0,
      "step": 22790
    },
    {
      "entropy": 5.057159185409546,
      "epoch": 1.9040260608085533,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046360786385623854,
      "loss": 4.7503,
      "mean_token_accuracy": 0.22089744061231614,
      "num_tokens": 41940833.0,
      "step": 22795
    },
    {
      "entropy": 5.095718622207642,
      "epoch": 1.9044437019712663,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004635916293785015,
      "loss": 4.8492,
      "mean_token_accuracy": 0.21407412439584733,
      "num_tokens": 41949756.0,
      "step": 22800
    },
    {
      "entropy": 5.1392968654632565,
      "epoch": 1.9048613431339794,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004635753915992409,
      "loss": 4.9884,
      "mean_token_accuracy": 0.21084757149219513,
      "num_tokens": 41959447.0,
      "step": 22805
    },
    {
      "entropy": 5.177515363693237,
      "epoch": 1.9052789842966922,
      "grad_norm": 1.203125,
      "learning_rate": 0.000463559150518741,
      "loss": 4.879,
      "mean_token_accuracy": 0.20744201838970183,
      "num_tokens": 41968308.0,
      "step": 22810
    },
    {
      "entropy": 5.117140531539917,
      "epoch": 1.9056966254594054,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004635429061372864,
      "loss": 4.8176,
      "mean_token_accuracy": 0.2159143716096878,
      "num_tokens": 41976331.0,
      "step": 22815
    },
    {
      "entropy": 5.119119119644165,
      "epoch": 1.9061142666221182,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046352665845516113,
      "loss": 4.89,
      "mean_token_accuracy": 0.20784018337726592,
      "num_tokens": 41985386.0,
      "step": 22820
    },
    {
      "entropy": 5.065849542617798,
      "epoch": 1.9065319077848313,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004635104074726498,
      "loss": 4.7463,
      "mean_token_accuracy": 0.22000790983438492,
      "num_tokens": 41993517.0,
      "step": 22825
    },
    {
      "entropy": 4.969551277160645,
      "epoch": 1.9069495489475443,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004634941531900369,
      "loss": 4.7539,
      "mean_token_accuracy": 0.21360609382390977,
      "num_tokens": 42002624.0,
      "step": 22830
    },
    {
      "entropy": 5.033498334884643,
      "epoch": 1.9073671901102571,
      "grad_norm": 1.3125,
      "learning_rate": 0.00046347789560760696,
      "loss": 4.8321,
      "mean_token_accuracy": 0.21534405052661895,
      "num_tokens": 42012128.0,
      "step": 22835
    },
    {
      "entropy": 5.071072196960449,
      "epoch": 1.9077848312729704,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004634616347256446,
      "loss": 4.7388,
      "mean_token_accuracy": 0.22186629623174667,
      "num_tokens": 42021243.0,
      "step": 22840
    },
    {
      "entropy": 5.097636556625366,
      "epoch": 1.9082024724356832,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046344537054443443,
      "loss": 4.8946,
      "mean_token_accuracy": 0.2124026522040367,
      "num_tokens": 42031397.0,
      "step": 22845
    },
    {
      "entropy": 5.203726673126221,
      "epoch": 1.9086201135983962,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004634291030642613,
      "loss": 4.8697,
      "mean_token_accuracy": 0.2152503713965416,
      "num_tokens": 42040560.0,
      "step": 22850
    },
    {
      "entropy": 5.081666374206543,
      "epoch": 1.9090377547611093,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004634128322854098,
      "loss": 4.8183,
      "mean_token_accuracy": 0.21716260761022568,
      "num_tokens": 42048824.0,
      "step": 22855
    },
    {
      "entropy": 5.053449869155884,
      "epoch": 1.9094553959238223,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004633965582081649,
      "loss": 4.8472,
      "mean_token_accuracy": 0.21499510407447814,
      "num_tokens": 42057460.0,
      "step": 22860
    },
    {
      "entropy": 5.0824875831604,
      "epoch": 1.9098730370865353,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046338028083281154,
      "loss": 4.8542,
      "mean_token_accuracy": 0.21074575930833817,
      "num_tokens": 42066829.0,
      "step": 22865
    },
    {
      "entropy": 5.1877013683319095,
      "epoch": 1.9102906782492481,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046336400015963453,
      "loss": 4.924,
      "mean_token_accuracy": 0.20907529294490815,
      "num_tokens": 42075518.0,
      "step": 22870
    },
    {
      "entropy": 5.084411334991455,
      "epoch": 1.9107083194119614,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000463347716188919,
      "loss": 4.7944,
      "mean_token_accuracy": 0.21213047802448273,
      "num_tokens": 42083939.0,
      "step": 22875
    },
    {
      "entropy": 4.986818075180054,
      "epoch": 1.9111259605746742,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046333142892094996,
      "loss": 4.7608,
      "mean_token_accuracy": 0.22083452343940735,
      "num_tokens": 42092229.0,
      "step": 22880
    },
    {
      "entropy": 5.076681137084961,
      "epoch": 1.9115436017373872,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046331513835601263,
      "loss": 4.8333,
      "mean_token_accuracy": 0.2101759672164917,
      "num_tokens": 42101876.0,
      "step": 22885
    },
    {
      "entropy": 5.121485948562622,
      "epoch": 1.9119612429001003,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046329884449439207,
      "loss": 4.8498,
      "mean_token_accuracy": 0.2149060994386673,
      "num_tokens": 42110710.0,
      "step": 22890
    },
    {
      "entropy": 5.181482982635498,
      "epoch": 1.912378884062813,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046328254733637357,
      "loss": 4.9883,
      "mean_token_accuracy": 0.20243272930383682,
      "num_tokens": 42120516.0,
      "step": 22895
    },
    {
      "entropy": 5.104342317581176,
      "epoch": 1.9127965252255263,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004632662468822425,
      "loss": 4.8129,
      "mean_token_accuracy": 0.2155335783958435,
      "num_tokens": 42129030.0,
      "step": 22900
    },
    {
      "entropy": 5.117251920700073,
      "epoch": 1.9132141663882392,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004632499431322842,
      "loss": 4.9187,
      "mean_token_accuracy": 0.21186275780200958,
      "num_tokens": 42138107.0,
      "step": 22905
    },
    {
      "entropy": 5.186989736557007,
      "epoch": 1.9136318075509522,
      "grad_norm": 1.203125,
      "learning_rate": 0.000463233636086784,
      "loss": 4.9894,
      "mean_token_accuracy": 0.20662340819835662,
      "num_tokens": 42147584.0,
      "step": 22910
    },
    {
      "entropy": 5.130022954940796,
      "epoch": 1.9140494487136652,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00046321732574602735,
      "loss": 4.7884,
      "mean_token_accuracy": 0.21458164751529693,
      "num_tokens": 42156751.0,
      "step": 22915
    },
    {
      "entropy": 5.072762680053711,
      "epoch": 1.9144670898763783,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046320101211029996,
      "loss": 4.7322,
      "mean_token_accuracy": 0.22304314076900483,
      "num_tokens": 42165204.0,
      "step": 22920
    },
    {
      "entropy": 5.069911766052246,
      "epoch": 1.9148847310390913,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046318469517988724,
      "loss": 4.8724,
      "mean_token_accuracy": 0.21034101843833924,
      "num_tokens": 42174028.0,
      "step": 22925
    },
    {
      "entropy": 5.098072671890259,
      "epoch": 1.915302372201804,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000463168374955075,
      "loss": 4.8684,
      "mean_token_accuracy": 0.21909199357032777,
      "num_tokens": 42183658.0,
      "step": 22930
    },
    {
      "entropy": 5.175391340255738,
      "epoch": 1.9157200133645174,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046315205143614873,
      "loss": 4.83,
      "mean_token_accuracy": 0.21352497041225432,
      "num_tokens": 42192509.0,
      "step": 22935
    },
    {
      "entropy": 5.1234039783477785,
      "epoch": 1.9161376545272302,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046313572462339437,
      "loss": 4.9019,
      "mean_token_accuracy": 0.20698396265506744,
      "num_tokens": 42201919.0,
      "step": 22940
    },
    {
      "entropy": 5.102031421661377,
      "epoch": 1.9165552956899432,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046311939451709776,
      "loss": 4.8803,
      "mean_token_accuracy": 0.21108624190092087,
      "num_tokens": 42211027.0,
      "step": 22945
    },
    {
      "entropy": 5.219144487380982,
      "epoch": 1.9169729368526562,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004631030611175447,
      "loss": 4.905,
      "mean_token_accuracy": 0.20949796587228775,
      "num_tokens": 42219677.0,
      "step": 22950
    },
    {
      "entropy": 5.18968825340271,
      "epoch": 1.917390578015369,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046308672442502103,
      "loss": 4.9231,
      "mean_token_accuracy": 0.20702987909317017,
      "num_tokens": 42228602.0,
      "step": 22955
    },
    {
      "entropy": 5.131395721435547,
      "epoch": 1.9178082191780823,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000463070384439813,
      "loss": 4.8471,
      "mean_token_accuracy": 0.2134843423962593,
      "num_tokens": 42238238.0,
      "step": 22960
    },
    {
      "entropy": 5.129674339294434,
      "epoch": 1.9182258603407951,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004630540411622064,
      "loss": 4.9385,
      "mean_token_accuracy": 0.2145329922437668,
      "num_tokens": 42247499.0,
      "step": 22965
    },
    {
      "entropy": 5.070890998840332,
      "epoch": 1.9186435015035082,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004630376945924875,
      "loss": 4.8089,
      "mean_token_accuracy": 0.22346198111772536,
      "num_tokens": 42256761.0,
      "step": 22970
    },
    {
      "entropy": 5.152292490005493,
      "epoch": 1.9190611426662212,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004630213447309424,
      "loss": 4.9515,
      "mean_token_accuracy": 0.20813161581754686,
      "num_tokens": 42266062.0,
      "step": 22975
    },
    {
      "entropy": 5.102982854843139,
      "epoch": 1.9194787838289342,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00046300499157785725,
      "loss": 4.7814,
      "mean_token_accuracy": 0.2129215955734253,
      "num_tokens": 42275510.0,
      "step": 22980
    },
    {
      "entropy": 5.147590827941895,
      "epoch": 1.9198964249916473,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046298863513351854,
      "loss": 4.9657,
      "mean_token_accuracy": 0.20080605745315552,
      "num_tokens": 42284905.0,
      "step": 22985
    },
    {
      "entropy": 5.112037467956543,
      "epoch": 1.92031406615436,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046297227539821245,
      "loss": 4.8698,
      "mean_token_accuracy": 0.2178536146879196,
      "num_tokens": 42294730.0,
      "step": 22990
    },
    {
      "entropy": 5.157242965698242,
      "epoch": 1.9207317073170733,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004629559123722254,
      "loss": 4.9059,
      "mean_token_accuracy": 0.2172243759036064,
      "num_tokens": 42304634.0,
      "step": 22995
    },
    {
      "entropy": 5.059633398056031,
      "epoch": 1.9211493484797861,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00046293954605584393,
      "loss": 4.7686,
      "mean_token_accuracy": 0.215757155418396,
      "num_tokens": 42313232.0,
      "step": 23000
    },
    {
      "entropy": 5.0564909934997555,
      "epoch": 1.9215669896424992,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004629231764493544,
      "loss": 4.8647,
      "mean_token_accuracy": 0.21533130407333373,
      "num_tokens": 42322320.0,
      "step": 23005
    },
    {
      "entropy": 5.127397441864014,
      "epoch": 1.9219846308052122,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004629068035530435,
      "loss": 4.789,
      "mean_token_accuracy": 0.21541272699832917,
      "num_tokens": 42331424.0,
      "step": 23010
    },
    {
      "entropy": 5.060230875015259,
      "epoch": 1.922402271967925,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004628904273671978,
      "loss": 4.8311,
      "mean_token_accuracy": 0.20472324192523955,
      "num_tokens": 42341759.0,
      "step": 23015
    },
    {
      "entropy": 5.0389186382293705,
      "epoch": 1.9228199131306383,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004628740478921041,
      "loss": 4.8322,
      "mean_token_accuracy": 0.21765658259391785,
      "num_tokens": 42351000.0,
      "step": 23020
    },
    {
      "entropy": 5.1297194480896,
      "epoch": 1.923237554293351,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046285766512804894,
      "loss": 4.8327,
      "mean_token_accuracy": 0.2199959546327591,
      "num_tokens": 42360797.0,
      "step": 23025
    },
    {
      "entropy": 5.040370130538941,
      "epoch": 1.9236551954560641,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00046284127907531925,
      "loss": 4.7411,
      "mean_token_accuracy": 0.22248944342136384,
      "num_tokens": 42370067.0,
      "step": 23030
    },
    {
      "entropy": 5.092399787902832,
      "epoch": 1.9240728366187771,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046282488973420187,
      "loss": 4.8253,
      "mean_token_accuracy": 0.21254073530435563,
      "num_tokens": 42379857.0,
      "step": 23035
    },
    {
      "entropy": 5.134926080703735,
      "epoch": 1.9244904777814902,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004628084971049837,
      "loss": 4.7887,
      "mean_token_accuracy": 0.21917556673288346,
      "num_tokens": 42388776.0,
      "step": 23040
    },
    {
      "entropy": 5.007963514328003,
      "epoch": 1.9249081189442032,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046279210118795175,
      "loss": 4.7438,
      "mean_token_accuracy": 0.22488941997289658,
      "num_tokens": 42397362.0,
      "step": 23045
    },
    {
      "entropy": 5.067755937576294,
      "epoch": 1.925325760106916,
      "grad_norm": 1.1875,
      "learning_rate": 0.000462775701983393,
      "loss": 4.8307,
      "mean_token_accuracy": 0.21096458286046982,
      "num_tokens": 42405856.0,
      "step": 23050
    },
    {
      "entropy": 5.168609952926635,
      "epoch": 1.9257434012696293,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004627592994915946,
      "loss": 4.9262,
      "mean_token_accuracy": 0.213077375292778,
      "num_tokens": 42415340.0,
      "step": 23055
    },
    {
      "entropy": 5.166008377075196,
      "epoch": 1.926161042432342,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046274289371284357,
      "loss": 4.8409,
      "mean_token_accuracy": 0.22310825139284135,
      "num_tokens": 42424113.0,
      "step": 23060
    },
    {
      "entropy": 5.057022571563721,
      "epoch": 1.9265786835950551,
      "grad_norm": 1.125,
      "learning_rate": 0.00046272648464742733,
      "loss": 4.7379,
      "mean_token_accuracy": 0.21822578758001326,
      "num_tokens": 42433833.0,
      "step": 23065
    },
    {
      "entropy": 5.009129619598388,
      "epoch": 1.9269963247577682,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004627100722956329,
      "loss": 4.74,
      "mean_token_accuracy": 0.22191323935985566,
      "num_tokens": 42442652.0,
      "step": 23070
    },
    {
      "entropy": 5.043701171875,
      "epoch": 1.927413965920481,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046269365665774773,
      "loss": 4.8673,
      "mean_token_accuracy": 0.21528924852609635,
      "num_tokens": 42452642.0,
      "step": 23075
    },
    {
      "entropy": 5.134158420562744,
      "epoch": 1.9278316070831942,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00046267723773405915,
      "loss": 4.8095,
      "mean_token_accuracy": 0.2176905706524849,
      "num_tokens": 42460623.0,
      "step": 23080
    },
    {
      "entropy": 5.028483152389526,
      "epoch": 1.928249248245907,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046266081552485467,
      "loss": 4.8087,
      "mean_token_accuracy": 0.2150767520070076,
      "num_tokens": 42469963.0,
      "step": 23085
    },
    {
      "entropy": 5.1627123832702635,
      "epoch": 1.92866688940862,
      "grad_norm": 1.125,
      "learning_rate": 0.00046264439003042176,
      "loss": 4.9263,
      "mean_token_accuracy": 0.21438873261213304,
      "num_tokens": 42479375.0,
      "step": 23090
    },
    {
      "entropy": 5.115921688079834,
      "epoch": 1.9290845305713331,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046262796125104785,
      "loss": 4.7611,
      "mean_token_accuracy": 0.21952075362205506,
      "num_tokens": 42487837.0,
      "step": 23095
    },
    {
      "entropy": 5.131635570526123,
      "epoch": 1.9295021717340461,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004626115291870206,
      "loss": 4.8948,
      "mean_token_accuracy": 0.21269350796937941,
      "num_tokens": 42496666.0,
      "step": 23100
    },
    {
      "entropy": 5.109059619903564,
      "epoch": 1.9299198128967592,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046259509383862784,
      "loss": 4.8538,
      "mean_token_accuracy": 0.2110302746295929,
      "num_tokens": 42507195.0,
      "step": 23105
    },
    {
      "entropy": 5.103094100952148,
      "epoch": 1.930337454059472,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004625786552061571,
      "loss": 4.7576,
      "mean_token_accuracy": 0.21754721105098723,
      "num_tokens": 42515879.0,
      "step": 23110
    },
    {
      "entropy": 5.145493888854981,
      "epoch": 1.9307550952221852,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004625622132898962,
      "loss": 4.9357,
      "mean_token_accuracy": 0.2125361055135727,
      "num_tokens": 42524877.0,
      "step": 23115
    },
    {
      "entropy": 5.027052116394043,
      "epoch": 1.931172736384898,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046254576809013297,
      "loss": 4.8484,
      "mean_token_accuracy": 0.2161245286464691,
      "num_tokens": 42534019.0,
      "step": 23120
    },
    {
      "entropy": 5.194633102416992,
      "epoch": 1.931590377547611,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004625293196071554,
      "loss": 4.9233,
      "mean_token_accuracy": 0.21138359010219573,
      "num_tokens": 42543280.0,
      "step": 23125
    },
    {
      "entropy": 5.189955759048462,
      "epoch": 1.9320080187103241,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004625128678412513,
      "loss": 4.9459,
      "mean_token_accuracy": 0.20263389348983765,
      "num_tokens": 42553476.0,
      "step": 23130
    },
    {
      "entropy": 5.088600444793701,
      "epoch": 1.932425659873037,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046249641279270866,
      "loss": 4.855,
      "mean_token_accuracy": 0.21376865804195405,
      "num_tokens": 42562502.0,
      "step": 23135
    },
    {
      "entropy": 5.148070526123047,
      "epoch": 1.9328433010357502,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046247995446181575,
      "loss": 4.8192,
      "mean_token_accuracy": 0.21582479476928712,
      "num_tokens": 42570947.0,
      "step": 23140
    },
    {
      "entropy": 5.070112991333008,
      "epoch": 1.933260942198463,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004624634928488605,
      "loss": 4.7983,
      "mean_token_accuracy": 0.21887121349573135,
      "num_tokens": 42579548.0,
      "step": 23145
    },
    {
      "entropy": 5.025719547271729,
      "epoch": 1.933678583361176,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004624470279541312,
      "loss": 4.7879,
      "mean_token_accuracy": 0.20987868458032607,
      "num_tokens": 42588006.0,
      "step": 23150
    },
    {
      "entropy": 5.1715422630310055,
      "epoch": 1.934096224523889,
      "grad_norm": 1.25,
      "learning_rate": 0.00046243055977791597,
      "loss": 4.9206,
      "mean_token_accuracy": 0.21381058990955354,
      "num_tokens": 42597322.0,
      "step": 23155
    },
    {
      "entropy": 5.163667154312134,
      "epoch": 1.934513865686602,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004624140883205033,
      "loss": 4.8281,
      "mean_token_accuracy": 0.21329947263002397,
      "num_tokens": 42605942.0,
      "step": 23160
    },
    {
      "entropy": 5.091455459594727,
      "epoch": 1.9349315068493151,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046239761358218135,
      "loss": 4.9046,
      "mean_token_accuracy": 0.2111823722720146,
      "num_tokens": 42614870.0,
      "step": 23165
    },
    {
      "entropy": 5.118478775024414,
      "epoch": 1.935349148012028,
      "grad_norm": 1.125,
      "learning_rate": 0.0004623811355632386,
      "loss": 4.8478,
      "mean_token_accuracy": 0.21462835520505905,
      "num_tokens": 42624324.0,
      "step": 23170
    },
    {
      "entropy": 5.108498191833496,
      "epoch": 1.9357667891747412,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046236465426396353,
      "loss": 4.8556,
      "mean_token_accuracy": 0.2033085897564888,
      "num_tokens": 42634120.0,
      "step": 23175
    },
    {
      "entropy": 5.044377994537354,
      "epoch": 1.936184430337454,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004623481696846446,
      "loss": 4.7869,
      "mean_token_accuracy": 0.22186776697635652,
      "num_tokens": 42642709.0,
      "step": 23180
    },
    {
      "entropy": 5.100552129745483,
      "epoch": 1.936602071500167,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004623316818255706,
      "loss": 4.85,
      "mean_token_accuracy": 0.21003633737564087,
      "num_tokens": 42652286.0,
      "step": 23185
    },
    {
      "entropy": 5.048695993423462,
      "epoch": 1.93701971266288,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004623151906870299,
      "loss": 4.816,
      "mean_token_accuracy": 0.21697536408901213,
      "num_tokens": 42662195.0,
      "step": 23190
    },
    {
      "entropy": 5.1263988494873045,
      "epoch": 1.937437353825593,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004622986962693113,
      "loss": 4.8417,
      "mean_token_accuracy": 0.21173581779003142,
      "num_tokens": 42671853.0,
      "step": 23195
    },
    {
      "entropy": 5.135759782791138,
      "epoch": 1.9378549949883062,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004622821985727036,
      "loss": 4.86,
      "mean_token_accuracy": 0.20648924857378007,
      "num_tokens": 42681677.0,
      "step": 23200
    },
    {
      "entropy": 5.037672328948974,
      "epoch": 1.938272636151019,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004622656975974956,
      "loss": 4.7423,
      "mean_token_accuracy": 0.2183698073029518,
      "num_tokens": 42690686.0,
      "step": 23205
    },
    {
      "entropy": 5.071998023986817,
      "epoch": 1.938690277313732,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046224919334397616,
      "loss": 4.8257,
      "mean_token_accuracy": 0.21545556038618088,
      "num_tokens": 42699967.0,
      "step": 23210
    },
    {
      "entropy": 5.095080423355102,
      "epoch": 1.939107918476445,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046223268581243416,
      "loss": 4.831,
      "mean_token_accuracy": 0.22006621062755585,
      "num_tokens": 42708185.0,
      "step": 23215
    },
    {
      "entropy": 5.076745796203613,
      "epoch": 1.939525559639158,
      "grad_norm": 1.125,
      "learning_rate": 0.00046221617500315866,
      "loss": 4.8492,
      "mean_token_accuracy": 0.2083833858370781,
      "num_tokens": 42717856.0,
      "step": 23220
    },
    {
      "entropy": 5.08005657196045,
      "epoch": 1.939943200801871,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046219966091643865,
      "loss": 4.8026,
      "mean_token_accuracy": 0.21742009818553926,
      "num_tokens": 42727036.0,
      "step": 23225
    },
    {
      "entropy": 5.133422613143921,
      "epoch": 1.940360841964584,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046218314355256324,
      "loss": 4.8325,
      "mean_token_accuracy": 0.22207117080688477,
      "num_tokens": 42736593.0,
      "step": 23230
    },
    {
      "entropy": 5.130580234527588,
      "epoch": 1.9407784831272972,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004621666229118216,
      "loss": 4.8291,
      "mean_token_accuracy": 0.22436176985502243,
      "num_tokens": 42745728.0,
      "step": 23235
    },
    {
      "entropy": 5.091387128829956,
      "epoch": 1.94119612429001,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046215009899450296,
      "loss": 4.8371,
      "mean_token_accuracy": 0.21804141253232956,
      "num_tokens": 42754638.0,
      "step": 23240
    },
    {
      "entropy": 5.074475574493408,
      "epoch": 1.941613765452723,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004621335718008965,
      "loss": 4.765,
      "mean_token_accuracy": 0.2168832004070282,
      "num_tokens": 42763474.0,
      "step": 23245
    },
    {
      "entropy": 5.08216290473938,
      "epoch": 1.942031406615436,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046211704133129164,
      "loss": 4.9154,
      "mean_token_accuracy": 0.21135921627283097,
      "num_tokens": 42772440.0,
      "step": 23250
    },
    {
      "entropy": 5.078321409225464,
      "epoch": 1.9424490477781489,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00046210050758597776,
      "loss": 4.7289,
      "mean_token_accuracy": 0.22576452940702438,
      "num_tokens": 42780369.0,
      "step": 23255
    },
    {
      "entropy": 5.056400442123413,
      "epoch": 1.9428666889408621,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046208397056524424,
      "loss": 4.7597,
      "mean_token_accuracy": 0.22056317925453187,
      "num_tokens": 42789652.0,
      "step": 23260
    },
    {
      "entropy": 5.130565404891968,
      "epoch": 1.943284330103575,
      "grad_norm": 1.125,
      "learning_rate": 0.0004620674302693807,
      "loss": 4.9019,
      "mean_token_accuracy": 0.206977479159832,
      "num_tokens": 42798067.0,
      "step": 23265
    },
    {
      "entropy": 5.099612665176392,
      "epoch": 1.943701971266288,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004620508866986765,
      "loss": 4.7753,
      "mean_token_accuracy": 0.21949685364961624,
      "num_tokens": 42806010.0,
      "step": 23270
    },
    {
      "entropy": 5.106219530105591,
      "epoch": 1.944119612429001,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046203433985342145,
      "loss": 4.8474,
      "mean_token_accuracy": 0.21058331727981566,
      "num_tokens": 42815596.0,
      "step": 23275
    },
    {
      "entropy": 5.0715056419372555,
      "epoch": 1.944537253591714,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004620177897339051,
      "loss": 4.7841,
      "mean_token_accuracy": 0.21333690583705903,
      "num_tokens": 42824953.0,
      "step": 23280
    },
    {
      "entropy": 5.1172278881072994,
      "epoch": 1.944954894754427,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004620012363404173,
      "loss": 4.9172,
      "mean_token_accuracy": 0.2116819754242897,
      "num_tokens": 42833737.0,
      "step": 23285
    },
    {
      "entropy": 5.173597908020019,
      "epoch": 1.9453725359171399,
      "grad_norm": 1.125,
      "learning_rate": 0.00046198467967324783,
      "loss": 4.9271,
      "mean_token_accuracy": 0.20872870832681656,
      "num_tokens": 42843674.0,
      "step": 23290
    },
    {
      "entropy": 5.058358001708984,
      "epoch": 1.9457901770798531,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004619681197326863,
      "loss": 4.7246,
      "mean_token_accuracy": 0.22620370090007783,
      "num_tokens": 42852510.0,
      "step": 23295
    },
    {
      "entropy": 4.9518805027008055,
      "epoch": 1.946207818242566,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004619515565190229,
      "loss": 4.6637,
      "mean_token_accuracy": 0.2301870107650757,
      "num_tokens": 42860482.0,
      "step": 23300
    },
    {
      "entropy": 5.08648157119751,
      "epoch": 1.946625459405279,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046193499003254737,
      "loss": 4.8413,
      "mean_token_accuracy": 0.2134243443608284,
      "num_tokens": 42869413.0,
      "step": 23305
    },
    {
      "entropy": 5.185434007644654,
      "epoch": 1.947043100567992,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004619184202735499,
      "loss": 4.9655,
      "mean_token_accuracy": 0.20856055319309236,
      "num_tokens": 42879032.0,
      "step": 23310
    },
    {
      "entropy": 5.124244689941406,
      "epoch": 1.9474607417307048,
      "grad_norm": 1.25,
      "learning_rate": 0.0004619018472423205,
      "loss": 4.8037,
      "mean_token_accuracy": 0.2155033081769943,
      "num_tokens": 42887989.0,
      "step": 23315
    },
    {
      "entropy": 5.116862773895264,
      "epoch": 1.947878382893418,
      "grad_norm": 1.265625,
      "learning_rate": 0.00046188527093914925,
      "loss": 4.8882,
      "mean_token_accuracy": 0.20928671211004257,
      "num_tokens": 42897268.0,
      "step": 23320
    },
    {
      "entropy": 5.113129377365112,
      "epoch": 1.948296024056131,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004618686913643264,
      "loss": 4.8093,
      "mean_token_accuracy": 0.21528707593679428,
      "num_tokens": 42906186.0,
      "step": 23325
    },
    {
      "entropy": 5.010228538513184,
      "epoch": 1.948713665218844,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004618521085181421,
      "loss": 4.7693,
      "mean_token_accuracy": 0.22311636209487914,
      "num_tokens": 42915457.0,
      "step": 23330
    },
    {
      "entropy": 5.161957311630249,
      "epoch": 1.949131306381557,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046183552240088683,
      "loss": 4.9237,
      "mean_token_accuracy": 0.21166706085205078,
      "num_tokens": 42925848.0,
      "step": 23335
    },
    {
      "entropy": 5.121760559082031,
      "epoch": 1.94954894754427,
      "grad_norm": 1.265625,
      "learning_rate": 0.00046181893301285075,
      "loss": 4.8971,
      "mean_token_accuracy": 0.22164395600557327,
      "num_tokens": 42936084.0,
      "step": 23340
    },
    {
      "entropy": 5.0221185207366945,
      "epoch": 1.949966588706983,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004618023403543245,
      "loss": 4.7912,
      "mean_token_accuracy": 0.22692776769399642,
      "num_tokens": 42945303.0,
      "step": 23345
    },
    {
      "entropy": 5.002455711364746,
      "epoch": 1.9503842298696958,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046178574442559835,
      "loss": 4.8438,
      "mean_token_accuracy": 0.21389489620923996,
      "num_tokens": 42955144.0,
      "step": 23350
    },
    {
      "entropy": 5.121132516860962,
      "epoch": 1.950801871032409,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004617691452269629,
      "loss": 4.7913,
      "mean_token_accuracy": 0.21731430292129517,
      "num_tokens": 42964492.0,
      "step": 23355
    },
    {
      "entropy": 5.021245765686035,
      "epoch": 1.951219512195122,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004617525427587087,
      "loss": 4.835,
      "mean_token_accuracy": 0.2203925982117653,
      "num_tokens": 42973731.0,
      "step": 23360
    },
    {
      "entropy": 4.941076755523682,
      "epoch": 1.951637153357835,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046173593702112655,
      "loss": 4.7174,
      "mean_token_accuracy": 0.2356455519795418,
      "num_tokens": 42981892.0,
      "step": 23365
    },
    {
      "entropy": 5.079247903823853,
      "epoch": 1.952054794520548,
      "grad_norm": 1.15625,
      "learning_rate": 0.000461719328014507,
      "loss": 4.7899,
      "mean_token_accuracy": 0.21950417160987853,
      "num_tokens": 42990877.0,
      "step": 23370
    },
    {
      "entropy": 5.224959897994995,
      "epoch": 1.9524724356832608,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046170271573914084,
      "loss": 4.9761,
      "mean_token_accuracy": 0.20623057931661606,
      "num_tokens": 43001371.0,
      "step": 23375
    },
    {
      "entropy": 5.0978532314300535,
      "epoch": 1.952890076845974,
      "grad_norm": 1.125,
      "learning_rate": 0.00046168610019531894,
      "loss": 4.7979,
      "mean_token_accuracy": 0.21921111643314362,
      "num_tokens": 43010651.0,
      "step": 23380
    },
    {
      "entropy": 5.062100744247436,
      "epoch": 1.9533077180086869,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004616694813833322,
      "loss": 4.8264,
      "mean_token_accuracy": 0.21819596886634826,
      "num_tokens": 43018883.0,
      "step": 23385
    },
    {
      "entropy": 5.113165998458863,
      "epoch": 1.9537253591714,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004616528593034713,
      "loss": 4.8241,
      "mean_token_accuracy": 0.21534452587366104,
      "num_tokens": 43027760.0,
      "step": 23390
    },
    {
      "entropy": 5.113671779632568,
      "epoch": 1.954143000334113,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004616362339560276,
      "loss": 4.7626,
      "mean_token_accuracy": 0.21767642349004745,
      "num_tokens": 43038227.0,
      "step": 23395
    },
    {
      "entropy": 4.986747789382934,
      "epoch": 1.954560641496826,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00046161960534129193,
      "loss": 4.6972,
      "mean_token_accuracy": 0.2262544095516205,
      "num_tokens": 43046792.0,
      "step": 23400
    },
    {
      "entropy": 5.048653936386108,
      "epoch": 1.954978282659539,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004616029734595555,
      "loss": 4.7893,
      "mean_token_accuracy": 0.21052355468273162,
      "num_tokens": 43055895.0,
      "step": 23405
    },
    {
      "entropy": 5.1109299659729,
      "epoch": 1.9553959238222518,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004615863383111093,
      "loss": 4.885,
      "mean_token_accuracy": 0.20687796622514726,
      "num_tokens": 43065631.0,
      "step": 23410
    },
    {
      "entropy": 5.191273784637451,
      "epoch": 1.955813564984965,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004615696998962447,
      "loss": 4.9531,
      "mean_token_accuracy": 0.21272904574871063,
      "num_tokens": 43074757.0,
      "step": 23415
    },
    {
      "entropy": 5.141025161743164,
      "epoch": 1.9562312061476779,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004615530582152529,
      "loss": 4.8952,
      "mean_token_accuracy": 0.21532817780971528,
      "num_tokens": 43084472.0,
      "step": 23420
    },
    {
      "entropy": 5.068595552444458,
      "epoch": 1.956648847310391,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004615364132684253,
      "loss": 4.8696,
      "mean_token_accuracy": 0.21064190119504927,
      "num_tokens": 43092726.0,
      "step": 23425
    },
    {
      "entropy": 5.111958312988281,
      "epoch": 1.957066488473104,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004615197650560532,
      "loss": 4.8068,
      "mean_token_accuracy": 0.21577346622943877,
      "num_tokens": 43101481.0,
      "step": 23430
    },
    {
      "entropy": 5.14975323677063,
      "epoch": 1.9574841296358167,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004615031135784282,
      "loss": 4.7963,
      "mean_token_accuracy": 0.2174262672662735,
      "num_tokens": 43110484.0,
      "step": 23435
    },
    {
      "entropy": 5.1328693389892575,
      "epoch": 1.95790177079853,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046148645883584155,
      "loss": 4.9695,
      "mean_token_accuracy": 0.21071464270353318,
      "num_tokens": 43121532.0,
      "step": 23440
    },
    {
      "entropy": 5.105296516418457,
      "epoch": 1.9583194119612428,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004614698008285851,
      "loss": 4.8035,
      "mean_token_accuracy": 0.22718264609575273,
      "num_tokens": 43130976.0,
      "step": 23445
    },
    {
      "entropy": 5.099675416946411,
      "epoch": 1.9587370531239559,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046145313955695027,
      "loss": 4.9152,
      "mean_token_accuracy": 0.21279991716146468,
      "num_tokens": 43139907.0,
      "step": 23450
    },
    {
      "entropy": 5.165708732604981,
      "epoch": 1.9591546942866689,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004614364750212288,
      "loss": 4.9327,
      "mean_token_accuracy": 0.21223603934049606,
      "num_tokens": 43149141.0,
      "step": 23455
    },
    {
      "entropy": 5.162325239181518,
      "epoch": 1.959572335449382,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046141980722171246,
      "loss": 4.898,
      "mean_token_accuracy": 0.21808901727199553,
      "num_tokens": 43158206.0,
      "step": 23460
    },
    {
      "entropy": 5.116401195526123,
      "epoch": 1.959989976612095,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046140313615869287,
      "loss": 4.9044,
      "mean_token_accuracy": 0.2141252115368843,
      "num_tokens": 43168365.0,
      "step": 23465
    },
    {
      "entropy": 4.9771017074584964,
      "epoch": 1.9604076177748078,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046138646183246213,
      "loss": 4.7152,
      "mean_token_accuracy": 0.2226218029856682,
      "num_tokens": 43176907.0,
      "step": 23470
    },
    {
      "entropy": 5.020198106765747,
      "epoch": 1.960825258937521,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000461369784243312,
      "loss": 4.7569,
      "mean_token_accuracy": 0.22340376824140548,
      "num_tokens": 43186978.0,
      "step": 23475
    },
    {
      "entropy": 5.06596531867981,
      "epoch": 1.9612429001002338,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046135310339153455,
      "loss": 4.8589,
      "mean_token_accuracy": 0.2154845491051674,
      "num_tokens": 43196502.0,
      "step": 23480
    },
    {
      "entropy": 5.206251621246338,
      "epoch": 1.9616605412629469,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004613364192774216,
      "loss": 4.9535,
      "mean_token_accuracy": 0.20291962921619416,
      "num_tokens": 43205555.0,
      "step": 23485
    },
    {
      "entropy": 5.099065065383911,
      "epoch": 1.96207818242566,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046131973190126536,
      "loss": 4.8027,
      "mean_token_accuracy": 0.21723734736442565,
      "num_tokens": 43214043.0,
      "step": 23490
    },
    {
      "entropy": 5.100044584274292,
      "epoch": 1.9624958235883727,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046130304126335795,
      "loss": 4.8994,
      "mean_token_accuracy": 0.20467457473278045,
      "num_tokens": 43223483.0,
      "step": 23495
    },
    {
      "entropy": 5.135451412200927,
      "epoch": 1.962913464751086,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004612863473639915,
      "loss": 4.8471,
      "mean_token_accuracy": 0.21937456130981445,
      "num_tokens": 43231962.0,
      "step": 23500
    },
    {
      "entropy": 5.096758031845093,
      "epoch": 1.9633311059137988,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046126965020345834,
      "loss": 4.8197,
      "mean_token_accuracy": 0.22116044759750367,
      "num_tokens": 43240532.0,
      "step": 23505
    },
    {
      "entropy": 5.028209495544433,
      "epoch": 1.9637487470765118,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004612529497820508,
      "loss": 4.7528,
      "mean_token_accuracy": 0.21730397045612335,
      "num_tokens": 43249867.0,
      "step": 23510
    },
    {
      "entropy": 5.109793281555175,
      "epoch": 1.9641663882392248,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004612362461000611,
      "loss": 4.9113,
      "mean_token_accuracy": 0.2122024491429329,
      "num_tokens": 43259393.0,
      "step": 23515
    },
    {
      "entropy": 5.27632884979248,
      "epoch": 1.9645840294019379,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046121953915778175,
      "loss": 4.935,
      "mean_token_accuracy": 0.21812038421630858,
      "num_tokens": 43269012.0,
      "step": 23520
    },
    {
      "entropy": 5.13637342453003,
      "epoch": 1.965001670564651,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004612028289555052,
      "loss": 4.884,
      "mean_token_accuracy": 0.21098053604364395,
      "num_tokens": 43278032.0,
      "step": 23525
    },
    {
      "entropy": 5.052732372283936,
      "epoch": 1.9654193117273637,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046118611549352407,
      "loss": 4.8617,
      "mean_token_accuracy": 0.21872226744890214,
      "num_tokens": 43287352.0,
      "step": 23530
    },
    {
      "entropy": 5.164127445220947,
      "epoch": 1.965836952890077,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004611693987721308,
      "loss": 4.9253,
      "mean_token_accuracy": 0.21092384457588195,
      "num_tokens": 43297048.0,
      "step": 23535
    },
    {
      "entropy": 5.0703747272491455,
      "epoch": 1.9662545940527898,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046115267879161823,
      "loss": 4.8075,
      "mean_token_accuracy": 0.22036059200763702,
      "num_tokens": 43306461.0,
      "step": 23540
    },
    {
      "entropy": 5.129823684692383,
      "epoch": 1.9666722352155028,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046113595555227885,
      "loss": 4.8377,
      "mean_token_accuracy": 0.20963527411222457,
      "num_tokens": 43315906.0,
      "step": 23545
    },
    {
      "entropy": 5.181406927108765,
      "epoch": 1.9670898763782159,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004611192290544055,
      "loss": 4.8932,
      "mean_token_accuracy": 0.2158416509628296,
      "num_tokens": 43325433.0,
      "step": 23550
    },
    {
      "entropy": 5.1412745952606205,
      "epoch": 1.9675075175409287,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00046110249929829104,
      "loss": 4.9162,
      "mean_token_accuracy": 0.21302515715360643,
      "num_tokens": 43333510.0,
      "step": 23555
    },
    {
      "entropy": 5.06655216217041,
      "epoch": 1.967925158703642,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004610857662842283,
      "loss": 4.8071,
      "mean_token_accuracy": 0.21688029021024705,
      "num_tokens": 43342985.0,
      "step": 23560
    },
    {
      "entropy": 5.184947204589844,
      "epoch": 1.9683427998663547,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004610690300125103,
      "loss": 4.8328,
      "mean_token_accuracy": 0.21713825464248657,
      "num_tokens": 43352894.0,
      "step": 23565
    },
    {
      "entropy": 5.139061021804809,
      "epoch": 1.9687604410290678,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046105229048342994,
      "loss": 4.8739,
      "mean_token_accuracy": 0.21203415393829345,
      "num_tokens": 43362667.0,
      "step": 23570
    },
    {
      "entropy": 5.092211961746216,
      "epoch": 1.9691780821917808,
      "grad_norm": 1.125,
      "learning_rate": 0.00046103554769728026,
      "loss": 4.8309,
      "mean_token_accuracy": 0.21767971366643907,
      "num_tokens": 43372212.0,
      "step": 23575
    },
    {
      "entropy": 5.069557952880859,
      "epoch": 1.9695957233544938,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004610188016543544,
      "loss": 4.8035,
      "mean_token_accuracy": 0.22120688557624818,
      "num_tokens": 43381958.0,
      "step": 23580
    },
    {
      "entropy": 5.122724390029907,
      "epoch": 1.9700133645172069,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046100205235494554,
      "loss": 4.8707,
      "mean_token_accuracy": 0.20878307968378068,
      "num_tokens": 43391641.0,
      "step": 23585
    },
    {
      "entropy": 5.164703512191773,
      "epoch": 1.9704310056799197,
      "grad_norm": 1.265625,
      "learning_rate": 0.00046098529979934677,
      "loss": 4.8625,
      "mean_token_accuracy": 0.21198333352804183,
      "num_tokens": 43401261.0,
      "step": 23590
    },
    {
      "entropy": 5.054232835769653,
      "epoch": 1.970848646842633,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004609685439878515,
      "loss": 4.7492,
      "mean_token_accuracy": 0.213791361451149,
      "num_tokens": 43409829.0,
      "step": 23595
    },
    {
      "entropy": 5.030879783630371,
      "epoch": 1.9712662880053458,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000460951784920753,
      "loss": 4.7541,
      "mean_token_accuracy": 0.22707380056381227,
      "num_tokens": 43418536.0,
      "step": 23600
    },
    {
      "entropy": 5.046954154968262,
      "epoch": 1.9716839291680588,
      "grad_norm": 1.25,
      "learning_rate": 0.0004609350225983447,
      "loss": 4.744,
      "mean_token_accuracy": 0.22147369384765625,
      "num_tokens": 43427667.0,
      "step": 23605
    },
    {
      "entropy": 5.068442058563233,
      "epoch": 1.9721015703307718,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046091825702092,
      "loss": 4.7952,
      "mean_token_accuracy": 0.21707571744918824,
      "num_tokens": 43436986.0,
      "step": 23610
    },
    {
      "entropy": 5.124598360061645,
      "epoch": 1.9725192114934846,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046090148818877244,
      "loss": 4.9367,
      "mean_token_accuracy": 0.2065209373831749,
      "num_tokens": 43446969.0,
      "step": 23615
    },
    {
      "entropy": 5.156751012802124,
      "epoch": 1.972936852656198,
      "grad_norm": 1.125,
      "learning_rate": 0.0004608847161021955,
      "loss": 4.9387,
      "mean_token_accuracy": 0.21278195679187775,
      "num_tokens": 43456160.0,
      "step": 23620
    },
    {
      "entropy": 5.200597858428955,
      "epoch": 1.9733544938189107,
      "grad_norm": 1.203125,
      "learning_rate": 0.000460867940761483,
      "loss": 4.9226,
      "mean_token_accuracy": 0.2115064725279808,
      "num_tokens": 43465767.0,
      "step": 23625
    },
    {
      "entropy": 5.160815906524658,
      "epoch": 1.9737721349816237,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004608511621669283,
      "loss": 4.8875,
      "mean_token_accuracy": 0.21367149502038957,
      "num_tokens": 43474493.0,
      "step": 23630
    },
    {
      "entropy": 5.077327489852905,
      "epoch": 1.9741897761443368,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046083438031882534,
      "loss": 4.8597,
      "mean_token_accuracy": 0.21410962641239167,
      "num_tokens": 43483410.0,
      "step": 23635
    },
    {
      "entropy": 5.054902458190918,
      "epoch": 1.9746074173070498,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004608175952174679,
      "loss": 4.7569,
      "mean_token_accuracy": 0.22078813910484313,
      "num_tokens": 43492532.0,
      "step": 23640
    },
    {
      "entropy": 5.072205448150635,
      "epoch": 1.9750250584697628,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004608008068631498,
      "loss": 4.7757,
      "mean_token_accuracy": 0.22482618242502211,
      "num_tokens": 43501282.0,
      "step": 23645
    },
    {
      "entropy": 5.104024267196655,
      "epoch": 1.9754426996324757,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046078401525616493,
      "loss": 4.8178,
      "mean_token_accuracy": 0.21110233068466186,
      "num_tokens": 43510355.0,
      "step": 23650
    },
    {
      "entropy": 5.0458978652954105,
      "epoch": 1.975860340795189,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004607672203968072,
      "loss": 4.8149,
      "mean_token_accuracy": 0.2232624590396881,
      "num_tokens": 43519179.0,
      "step": 23655
    },
    {
      "entropy": 5.093927240371704,
      "epoch": 1.9762779819579017,
      "grad_norm": 1.125,
      "learning_rate": 0.00046075042228537074,
      "loss": 4.8598,
      "mean_token_accuracy": 0.21129736602306365,
      "num_tokens": 43528306.0,
      "step": 23660
    },
    {
      "entropy": 5.244199180603028,
      "epoch": 1.9766956231206148,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004607336209221494,
      "loss": 4.9734,
      "mean_token_accuracy": 0.20802890807390212,
      "num_tokens": 43537075.0,
      "step": 23665
    },
    {
      "entropy": 5.134850549697876,
      "epoch": 1.9771132642833278,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004607168163074376,
      "loss": 4.8564,
      "mean_token_accuracy": 0.2178254932165146,
      "num_tokens": 43545356.0,
      "step": 23670
    },
    {
      "entropy": 5.132081413269043,
      "epoch": 1.9775309054460406,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004607000084415294,
      "loss": 4.8921,
      "mean_token_accuracy": 0.21406056880950927,
      "num_tokens": 43555012.0,
      "step": 23675
    },
    {
      "entropy": 5.1751494884490965,
      "epoch": 1.9779485466087539,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046068319732471894,
      "loss": 4.9312,
      "mean_token_accuracy": 0.20768446177244188,
      "num_tokens": 43565245.0,
      "step": 23680
    },
    {
      "entropy": 5.238878679275513,
      "epoch": 1.9783661877714667,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004606663829573006,
      "loss": 5.0035,
      "mean_token_accuracy": 0.20084784924983978,
      "num_tokens": 43574246.0,
      "step": 23685
    },
    {
      "entropy": 4.970189380645752,
      "epoch": 1.9787838289341797,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004606495653395688,
      "loss": 4.6439,
      "mean_token_accuracy": 0.23419489562511445,
      "num_tokens": 43582494.0,
      "step": 23690
    },
    {
      "entropy": 5.137055253982544,
      "epoch": 1.9792014700968927,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004606327444718179,
      "loss": 4.9155,
      "mean_token_accuracy": 0.2107292264699936,
      "num_tokens": 43593289.0,
      "step": 23695
    },
    {
      "entropy": 5.16360182762146,
      "epoch": 1.9796191112596058,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004606159203543423,
      "loss": 4.819,
      "mean_token_accuracy": 0.22009043246507645,
      "num_tokens": 43602062.0,
      "step": 23700
    },
    {
      "entropy": 5.146829414367676,
      "epoch": 1.9800367524223188,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004605990929874366,
      "loss": 4.8949,
      "mean_token_accuracy": 0.2104905754327774,
      "num_tokens": 43610830.0,
      "step": 23705
    },
    {
      "entropy": 5.1346289157867435,
      "epoch": 1.9804543935850316,
      "grad_norm": 1.125,
      "learning_rate": 0.0004605822623713954,
      "loss": 4.869,
      "mean_token_accuracy": 0.2065831184387207,
      "num_tokens": 43619656.0,
      "step": 23710
    },
    {
      "entropy": 5.1112961769104,
      "epoch": 1.9808720347477449,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046056542850651333,
      "loss": 4.8804,
      "mean_token_accuracy": 0.2186099484562874,
      "num_tokens": 43629727.0,
      "step": 23715
    },
    {
      "entropy": 5.120416355133057,
      "epoch": 1.9812896759104577,
      "grad_norm": 1.234375,
      "learning_rate": 0.000460548591393085,
      "loss": 4.9347,
      "mean_token_accuracy": 0.21768554598093032,
      "num_tokens": 43638532.0,
      "step": 23720
    },
    {
      "entropy": 5.1631354808807375,
      "epoch": 1.9817073170731707,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046053175103140533,
      "loss": 4.9175,
      "mean_token_accuracy": 0.21947928071022033,
      "num_tokens": 43647679.0,
      "step": 23725
    },
    {
      "entropy": 5.184066724777222,
      "epoch": 1.9821249582358837,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046051490742176894,
      "loss": 4.8153,
      "mean_token_accuracy": 0.21433355063199996,
      "num_tokens": 43656340.0,
      "step": 23730
    },
    {
      "entropy": 5.074196767807007,
      "epoch": 1.9825425993985966,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004604980605644708,
      "loss": 4.7818,
      "mean_token_accuracy": 0.2113337993621826,
      "num_tokens": 43665721.0,
      "step": 23735
    },
    {
      "entropy": 5.123082256317138,
      "epoch": 1.9829602405613098,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004604812104598059,
      "loss": 4.8505,
      "mean_token_accuracy": 0.21697208136320115,
      "num_tokens": 43674640.0,
      "step": 23740
    },
    {
      "entropy": 5.141788196563721,
      "epoch": 1.9833778817240226,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004604643571080691,
      "loss": 4.8582,
      "mean_token_accuracy": 0.21289218813180924,
      "num_tokens": 43683080.0,
      "step": 23745
    },
    {
      "entropy": 5.035746240615845,
      "epoch": 1.9837955228867357,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046044750050955546,
      "loss": 4.7692,
      "mean_token_accuracy": 0.22134897708892823,
      "num_tokens": 43691259.0,
      "step": 23750
    },
    {
      "entropy": 5.1049798965454105,
      "epoch": 1.9842131640494487,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046043064066456015,
      "loss": 4.8335,
      "mean_token_accuracy": 0.21959354728460312,
      "num_tokens": 43699881.0,
      "step": 23755
    },
    {
      "entropy": 5.068230819702149,
      "epoch": 1.9846308052121617,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046041377757337826,
      "loss": 4.8399,
      "mean_token_accuracy": 0.22023907899856568,
      "num_tokens": 43708742.0,
      "step": 23760
    },
    {
      "entropy": 5.124606704711914,
      "epoch": 1.9850484463748748,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046039691123630496,
      "loss": 4.8314,
      "mean_token_accuracy": 0.21102532595396042,
      "num_tokens": 43718450.0,
      "step": 23765
    },
    {
      "entropy": 5.116872882843017,
      "epoch": 1.9854660875375876,
      "grad_norm": 1.125,
      "learning_rate": 0.00046038004165363554,
      "loss": 4.7949,
      "mean_token_accuracy": 0.21418662518262863,
      "num_tokens": 43728340.0,
      "step": 23770
    },
    {
      "entropy": 5.0431640625,
      "epoch": 1.9858837287003008,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046036316882566545,
      "loss": 4.7571,
      "mean_token_accuracy": 0.2188951402902603,
      "num_tokens": 43737287.0,
      "step": 23775
    },
    {
      "entropy": 5.033590173721313,
      "epoch": 1.9863013698630136,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046034629275268997,
      "loss": 4.7645,
      "mean_token_accuracy": 0.22097592949867248,
      "num_tokens": 43745899.0,
      "step": 23780
    },
    {
      "entropy": 5.193128776550293,
      "epoch": 1.9867190110257267,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046032941343500443,
      "loss": 4.9047,
      "mean_token_accuracy": 0.20623429268598556,
      "num_tokens": 43755109.0,
      "step": 23785
    },
    {
      "entropy": 5.091243743896484,
      "epoch": 1.9871366521884397,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046031253087290455,
      "loss": 4.7412,
      "mean_token_accuracy": 0.2219506546854973,
      "num_tokens": 43764216.0,
      "step": 23790
    },
    {
      "entropy": 5.052247476577759,
      "epoch": 1.9875542933511525,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046029564506668567,
      "loss": 4.786,
      "mean_token_accuracy": 0.21440149396657943,
      "num_tokens": 43773267.0,
      "step": 23795
    },
    {
      "entropy": 5.197661304473877,
      "epoch": 1.9879719345138658,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046027875601664355,
      "loss": 4.9343,
      "mean_token_accuracy": 0.20929832458496095,
      "num_tokens": 43782558.0,
      "step": 23800
    },
    {
      "entropy": 5.19368166923523,
      "epoch": 1.9883895756765786,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004602618637230737,
      "loss": 4.8738,
      "mean_token_accuracy": 0.2064577043056488,
      "num_tokens": 43791861.0,
      "step": 23805
    },
    {
      "entropy": 5.081134891510009,
      "epoch": 1.9888072168392918,
      "grad_norm": 1.140625,
      "learning_rate": 0.000460244968186272,
      "loss": 4.7898,
      "mean_token_accuracy": 0.213768270611763,
      "num_tokens": 43801064.0,
      "step": 23810
    },
    {
      "entropy": 5.08175048828125,
      "epoch": 1.9892248580020047,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046022806940653415,
      "loss": 4.789,
      "mean_token_accuracy": 0.2245810940861702,
      "num_tokens": 43810159.0,
      "step": 23815
    },
    {
      "entropy": 5.115678119659424,
      "epoch": 1.9896424991647177,
      "grad_norm": 1.078125,
      "learning_rate": 0.000460211167384156,
      "loss": 4.8854,
      "mean_token_accuracy": 0.21405591070652008,
      "num_tokens": 43819056.0,
      "step": 23820
    },
    {
      "entropy": 5.156491279602051,
      "epoch": 1.9900601403274307,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046019426211943334,
      "loss": 4.8512,
      "mean_token_accuracy": 0.21841284185647963,
      "num_tokens": 43828024.0,
      "step": 23825
    },
    {
      "entropy": 5.092713737487793,
      "epoch": 1.9904777814901435,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004601773536126623,
      "loss": 4.7421,
      "mean_token_accuracy": 0.22554012835025788,
      "num_tokens": 43836915.0,
      "step": 23830
    },
    {
      "entropy": 5.0426520824432375,
      "epoch": 1.9908954226528568,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004601604418641387,
      "loss": 4.7957,
      "mean_token_accuracy": 0.22066776156425477,
      "num_tokens": 43845661.0,
      "step": 23835
    },
    {
      "entropy": 5.156585454940796,
      "epoch": 1.9913130638155696,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046014352687415866,
      "loss": 4.8997,
      "mean_token_accuracy": 0.20710911005735397,
      "num_tokens": 43854876.0,
      "step": 23840
    },
    {
      "entropy": 5.046794080734253,
      "epoch": 1.9917307049782826,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046012660864301846,
      "loss": 4.7903,
      "mean_token_accuracy": 0.222848142683506,
      "num_tokens": 43864750.0,
      "step": 23845
    },
    {
      "entropy": 5.183694887161255,
      "epoch": 1.9921483461409957,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046010968717101404,
      "loss": 4.8872,
      "mean_token_accuracy": 0.21013874262571336,
      "num_tokens": 43874889.0,
      "step": 23850
    },
    {
      "entropy": 5.102280235290527,
      "epoch": 1.9925659873037087,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046009276245844176,
      "loss": 4.8264,
      "mean_token_accuracy": 0.20903584510087966,
      "num_tokens": 43884462.0,
      "step": 23855
    },
    {
      "entropy": 5.073216676712036,
      "epoch": 1.9929836284664217,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004600758345055979,
      "loss": 4.8372,
      "mean_token_accuracy": 0.2113531306385994,
      "num_tokens": 43893739.0,
      "step": 23860
    },
    {
      "entropy": 5.152301359176636,
      "epoch": 1.9934012696291346,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046005890331277875,
      "loss": 4.8787,
      "mean_token_accuracy": 0.21508461683988572,
      "num_tokens": 43904172.0,
      "step": 23865
    },
    {
      "entropy": 5.159586668014526,
      "epoch": 1.9938189107918478,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004600419688802807,
      "loss": 4.8839,
      "mean_token_accuracy": 0.21087169498205185,
      "num_tokens": 43912824.0,
      "step": 23870
    },
    {
      "entropy": 5.07195348739624,
      "epoch": 1.9942365519545606,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046002503120840037,
      "loss": 4.7683,
      "mean_token_accuracy": 0.21791437715291978,
      "num_tokens": 43921704.0,
      "step": 23875
    },
    {
      "entropy": 5.116472673416138,
      "epoch": 1.9946541931172737,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000460008090297434,
      "loss": 4.8095,
      "mean_token_accuracy": 0.217141592502594,
      "num_tokens": 43930400.0,
      "step": 23880
    },
    {
      "entropy": 5.081533193588257,
      "epoch": 1.9950718342799867,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00045999114614767847,
      "loss": 4.7444,
      "mean_token_accuracy": 0.22064459919929505,
      "num_tokens": 43939623.0,
      "step": 23885
    },
    {
      "entropy": 5.087355661392212,
      "epoch": 1.9954894754426995,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004599741987594302,
      "loss": 4.8658,
      "mean_token_accuracy": 0.21631970256567,
      "num_tokens": 43948902.0,
      "step": 23890
    },
    {
      "entropy": 5.098429107666016,
      "epoch": 1.9959071166054128,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004599572481329859,
      "loss": 4.8451,
      "mean_token_accuracy": 0.2131129249930382,
      "num_tokens": 43958747.0,
      "step": 23895
    },
    {
      "entropy": 5.166495656967163,
      "epoch": 1.9963247577681256,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004599402942686423,
      "loss": 4.9389,
      "mean_token_accuracy": 0.20774906426668166,
      "num_tokens": 43968766.0,
      "step": 23900
    },
    {
      "entropy": 5.080013084411621,
      "epoch": 1.9967423989308386,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004599233371666963,
      "loss": 4.8141,
      "mean_token_accuracy": 0.21549088954925538,
      "num_tokens": 43978212.0,
      "step": 23905
    },
    {
      "entropy": 5.09298620223999,
      "epoch": 1.9971600400935516,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004599063768274447,
      "loss": 4.8292,
      "mean_token_accuracy": 0.21492539793252946,
      "num_tokens": 43987167.0,
      "step": 23910
    },
    {
      "entropy": 5.099967432022095,
      "epoch": 1.9975776812562647,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004598894132511843,
      "loss": 4.7565,
      "mean_token_accuracy": 0.21749995648860931,
      "num_tokens": 43995405.0,
      "step": 23915
    },
    {
      "entropy": 5.135137367248535,
      "epoch": 1.9979953224189777,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004598724464382123,
      "loss": 4.8532,
      "mean_token_accuracy": 0.21893750578165055,
      "num_tokens": 44004062.0,
      "step": 23920
    },
    {
      "entropy": 5.121805620193482,
      "epoch": 1.9984129635816905,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00045985547638882553,
      "loss": 4.8784,
      "mean_token_accuracy": 0.21527248173952102,
      "num_tokens": 44013911.0,
      "step": 23925
    },
    {
      "entropy": 5.091158199310303,
      "epoch": 1.9988306047444038,
      "grad_norm": 1.375,
      "learning_rate": 0.00045983850310332117,
      "loss": 4.8811,
      "mean_token_accuracy": 0.21107331961393355,
      "num_tokens": 44023366.0,
      "step": 23930
    },
    {
      "entropy": 5.019389009475708,
      "epoch": 1.9992482459071166,
      "grad_norm": 1.125,
      "learning_rate": 0.0004598215265819963,
      "loss": 4.7429,
      "mean_token_accuracy": 0.22437112182378768,
      "num_tokens": 44031996.0,
      "step": 23935
    },
    {
      "entropy": 5.079756402969361,
      "epoch": 1.9996658870698296,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004598045468251481,
      "loss": 4.7445,
      "mean_token_accuracy": 0.2205282613635063,
      "num_tokens": 44041474.0,
      "step": 23940
    },
    {
      "entropy": 5.120324897766113,
      "epoch": 2.0000835282325427,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004597875638330739,
      "loss": 4.7714,
      "mean_token_accuracy": 0.217340786755085,
      "num_tokens": 44050213.0,
      "step": 23945
    },
    {
      "entropy": 5.000110054016114,
      "epoch": 2.0005011693952555,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045977057760607096,
      "loss": 4.6319,
      "mean_token_accuracy": 0.2305361360311508,
      "num_tokens": 44058686.0,
      "step": 23950
    },
    {
      "entropy": 5.024531459808349,
      "epoch": 2.0009188105579687,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004597535881444366,
      "loss": 4.7767,
      "mean_token_accuracy": 0.21958464235067368,
      "num_tokens": 44068316.0,
      "step": 23955
    },
    {
      "entropy": 5.092236852645874,
      "epoch": 2.0013364517206815,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045973659544846834,
      "loss": 4.7719,
      "mean_token_accuracy": 0.22335862666368483,
      "num_tokens": 44077474.0,
      "step": 23960
    },
    {
      "entropy": 4.9685813903808596,
      "epoch": 2.001754092883395,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004597195995184637,
      "loss": 4.6488,
      "mean_token_accuracy": 0.23240904062986373,
      "num_tokens": 44086334.0,
      "step": 23965
    },
    {
      "entropy": 5.101002311706543,
      "epoch": 2.0021717340461076,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004597026003547201,
      "loss": 4.7868,
      "mean_token_accuracy": 0.210799703001976,
      "num_tokens": 44095543.0,
      "step": 23970
    },
    {
      "entropy": 5.1453088283538815,
      "epoch": 2.0025893752088204,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045968559795753493,
      "loss": 4.9126,
      "mean_token_accuracy": 0.2139233261346817,
      "num_tokens": 44105162.0,
      "step": 23975
    },
    {
      "entropy": 5.127979183197022,
      "epoch": 2.0030070163715337,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045966859232720626,
      "loss": 4.7599,
      "mean_token_accuracy": 0.21717125475406646,
      "num_tokens": 44114087.0,
      "step": 23980
    },
    {
      "entropy": 5.200289154052735,
      "epoch": 2.0034246575342465,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004596515834640316,
      "loss": 4.9311,
      "mean_token_accuracy": 0.21223047971725464,
      "num_tokens": 44123186.0,
      "step": 23985
    },
    {
      "entropy": 5.052420949935913,
      "epoch": 2.0038422986969597,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004596345713683086,
      "loss": 4.7648,
      "mean_token_accuracy": 0.22038111686706544,
      "num_tokens": 44132126.0,
      "step": 23990
    },
    {
      "entropy": 5.139699029922485,
      "epoch": 2.0042599398596725,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045961755604033536,
      "loss": 4.8639,
      "mean_token_accuracy": 0.20909689962863923,
      "num_tokens": 44141703.0,
      "step": 23995
    },
    {
      "entropy": 5.135097217559815,
      "epoch": 2.0046775810223854,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004596005374804094,
      "loss": 4.7448,
      "mean_token_accuracy": 0.21900465488433837,
      "num_tokens": 44150512.0,
      "step": 24000
    },
    {
      "epoch": 2.0046775810223854,
      "eval_entropy": 4.907006504424066,
      "eval_loss": 4.844655990600586,
      "eval_mean_token_accuracy": 0.22557730255484756,
      "eval_num_tokens": 44150512.0,
      "eval_runtime": 27.5101,
      "eval_samples_per_second": 1372.695,
      "eval_steps_per_second": 171.61,
      "step": 24000
    },
    {
      "entropy": 5.1373734951019285,
      "epoch": 2.0050952221850986,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045958351568882883,
      "loss": 4.7422,
      "mean_token_accuracy": 0.2165664628148079,
      "num_tokens": 44160209.0,
      "step": 24005
    },
    {
      "entropy": 5.076629304885865,
      "epoch": 2.0055128633478114,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004595664906658917,
      "loss": 4.7673,
      "mean_token_accuracy": 0.21859950572252274,
      "num_tokens": 44169005.0,
      "step": 24010
    },
    {
      "entropy": 5.069589376449585,
      "epoch": 2.0059305045105247,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045954946241189606,
      "loss": 4.8205,
      "mean_token_accuracy": 0.21375761330127716,
      "num_tokens": 44178750.0,
      "step": 24015
    },
    {
      "entropy": 5.141050338745117,
      "epoch": 2.0063481456732375,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045953243092713986,
      "loss": 4.8655,
      "mean_token_accuracy": 0.21098292171955108,
      "num_tokens": 44188268.0,
      "step": 24020
    },
    {
      "entropy": 5.087001991271973,
      "epoch": 2.0067657868359507,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045951539621192137,
      "loss": 4.7134,
      "mean_token_accuracy": 0.22285671681165695,
      "num_tokens": 44197599.0,
      "step": 24025
    },
    {
      "entropy": 5.115604066848755,
      "epoch": 2.0071834279986636,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004594983582665387,
      "loss": 4.9204,
      "mean_token_accuracy": 0.2121204987168312,
      "num_tokens": 44207971.0,
      "step": 24030
    },
    {
      "entropy": 5.150535011291504,
      "epoch": 2.0076010691613764,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004594813170912902,
      "loss": 4.8265,
      "mean_token_accuracy": 0.20834348499774932,
      "num_tokens": 44217387.0,
      "step": 24035
    },
    {
      "entropy": 5.107284688949585,
      "epoch": 2.0080187103240896,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045946427268647424,
      "loss": 4.7597,
      "mean_token_accuracy": 0.21740923523902894,
      "num_tokens": 44226103.0,
      "step": 24040
    },
    {
      "entropy": 5.070657968521118,
      "epoch": 2.0084363514868024,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045944722505238914,
      "loss": 4.7504,
      "mean_token_accuracy": 0.21563316881656647,
      "num_tokens": 44236081.0,
      "step": 24045
    },
    {
      "entropy": 5.123218631744384,
      "epoch": 2.0088539926495157,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045943017418933324,
      "loss": 4.791,
      "mean_token_accuracy": 0.21535901129245758,
      "num_tokens": 44245934.0,
      "step": 24050
    },
    {
      "entropy": 5.05701379776001,
      "epoch": 2.0092716338122285,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004594131200976053,
      "loss": 4.7654,
      "mean_token_accuracy": 0.2206251636147499,
      "num_tokens": 44255629.0,
      "step": 24055
    },
    {
      "entropy": 5.054419040679932,
      "epoch": 2.0096892749749413,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045939606277750356,
      "loss": 4.7092,
      "mean_token_accuracy": 0.22217195034027098,
      "num_tokens": 44264954.0,
      "step": 24060
    },
    {
      "entropy": 5.065571451187134,
      "epoch": 2.0101069161376546,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004593790022293269,
      "loss": 4.7162,
      "mean_token_accuracy": 0.22150208950042724,
      "num_tokens": 44273739.0,
      "step": 24065
    },
    {
      "entropy": 5.101988458633423,
      "epoch": 2.0105245573003674,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004593619384533738,
      "loss": 4.8782,
      "mean_token_accuracy": 0.21237437278032303,
      "num_tokens": 44283365.0,
      "step": 24070
    },
    {
      "entropy": 5.1111267566680905,
      "epoch": 2.0109421984630806,
      "grad_norm": 1.25,
      "learning_rate": 0.00045934487144994306,
      "loss": 4.7996,
      "mean_token_accuracy": 0.21499489545822142,
      "num_tokens": 44292257.0,
      "step": 24075
    },
    {
      "entropy": 5.249235773086548,
      "epoch": 2.0113598396257935,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045932780121933354,
      "loss": 4.9167,
      "mean_token_accuracy": 0.2139566257596016,
      "num_tokens": 44302539.0,
      "step": 24080
    },
    {
      "entropy": 5.016484880447388,
      "epoch": 2.0117774807885067,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004593107277618439,
      "loss": 4.7189,
      "mean_token_accuracy": 0.21747651398181916,
      "num_tokens": 44311812.0,
      "step": 24085
    },
    {
      "entropy": 4.954787731170654,
      "epoch": 2.0121951219512195,
      "grad_norm": 1.125,
      "learning_rate": 0.0004592936510777731,
      "loss": 4.7441,
      "mean_token_accuracy": 0.22231363803148269,
      "num_tokens": 44321015.0,
      "step": 24090
    },
    {
      "entropy": 5.201602792739868,
      "epoch": 2.0126127631139323,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045927657116742016,
      "loss": 4.8444,
      "mean_token_accuracy": 0.2082935094833374,
      "num_tokens": 44330066.0,
      "step": 24095
    },
    {
      "entropy": 5.186386013031006,
      "epoch": 2.0130304042766456,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000459259488031084,
      "loss": 4.8993,
      "mean_token_accuracy": 0.20985613763332367,
      "num_tokens": 44339716.0,
      "step": 24100
    },
    {
      "entropy": 5.03304615020752,
      "epoch": 2.0134480454393584,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004592424016690637,
      "loss": 4.6292,
      "mean_token_accuracy": 0.2221498280763626,
      "num_tokens": 44348340.0,
      "step": 24105
    },
    {
      "entropy": 5.00019645690918,
      "epoch": 2.0138656866020717,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045922531208165843,
      "loss": 4.7601,
      "mean_token_accuracy": 0.22318482547998428,
      "num_tokens": 44358045.0,
      "step": 24110
    },
    {
      "entropy": 5.1064496517181395,
      "epoch": 2.0142833277647845,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004592082192691673,
      "loss": 4.782,
      "mean_token_accuracy": 0.21923305988311767,
      "num_tokens": 44366962.0,
      "step": 24115
    },
    {
      "entropy": 5.033675622940064,
      "epoch": 2.0147009689274973,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045919112323188955,
      "loss": 4.6916,
      "mean_token_accuracy": 0.2244187757372856,
      "num_tokens": 44376178.0,
      "step": 24120
    },
    {
      "entropy": 5.097725963592529,
      "epoch": 2.0151186100902105,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004591740239701244,
      "loss": 4.7676,
      "mean_token_accuracy": 0.21855437904596328,
      "num_tokens": 44385271.0,
      "step": 24125
    },
    {
      "entropy": 5.107425212860107,
      "epoch": 2.0155362512529233,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045915692148417136,
      "loss": 4.7417,
      "mean_token_accuracy": 0.22219817191362382,
      "num_tokens": 44394900.0,
      "step": 24130
    },
    {
      "entropy": 4.993979740142822,
      "epoch": 2.0159538924156366,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004591398157743297,
      "loss": 4.7211,
      "mean_token_accuracy": 0.2271378755569458,
      "num_tokens": 44403812.0,
      "step": 24135
    },
    {
      "entropy": 5.066956377029419,
      "epoch": 2.0163715335783494,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045912270684089903,
      "loss": 4.7901,
      "mean_token_accuracy": 0.21822398155927658,
      "num_tokens": 44413301.0,
      "step": 24140
    },
    {
      "entropy": 5.074260234832764,
      "epoch": 2.0167891747410627,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004591055946841786,
      "loss": 4.7039,
      "mean_token_accuracy": 0.22305625975131987,
      "num_tokens": 44422334.0,
      "step": 24145
    },
    {
      "entropy": 5.108462619781494,
      "epoch": 2.0172068159037755,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045908847930446815,
      "loss": 5.0336,
      "mean_token_accuracy": 0.20834842473268508,
      "num_tokens": 44432914.0,
      "step": 24150
    },
    {
      "entropy": 5.1307745456695555,
      "epoch": 2.0176244570664883,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004590713607020674,
      "loss": 4.698,
      "mean_token_accuracy": 0.2303885117173195,
      "num_tokens": 44442372.0,
      "step": 24155
    },
    {
      "entropy": 5.083110094070435,
      "epoch": 2.0180420982292016,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045905423887727584,
      "loss": 4.681,
      "mean_token_accuracy": 0.22685081958770753,
      "num_tokens": 44450866.0,
      "step": 24160
    },
    {
      "entropy": 5.030089282989502,
      "epoch": 2.0184597393919144,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004590371138303933,
      "loss": 4.7695,
      "mean_token_accuracy": 0.22078623473644257,
      "num_tokens": 44460224.0,
      "step": 24165
    },
    {
      "entropy": 5.110888528823852,
      "epoch": 2.0188773805546276,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004590199855617195,
      "loss": 4.7371,
      "mean_token_accuracy": 0.22038561552762986,
      "num_tokens": 44469231.0,
      "step": 24170
    },
    {
      "entropy": 5.079917955398559,
      "epoch": 2.0192950217173404,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004590028540715544,
      "loss": 4.7208,
      "mean_token_accuracy": 0.21908824443817138,
      "num_tokens": 44478415.0,
      "step": 24175
    },
    {
      "entropy": 5.041001319885254,
      "epoch": 2.0197126628800532,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004589857193601978,
      "loss": 4.7756,
      "mean_token_accuracy": 0.22265952676534653,
      "num_tokens": 44486648.0,
      "step": 24180
    },
    {
      "entropy": 5.104247426986694,
      "epoch": 2.0201303040427665,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045896858142794976,
      "loss": 4.7795,
      "mean_token_accuracy": 0.21616672724485397,
      "num_tokens": 44495111.0,
      "step": 24185
    },
    {
      "entropy": 5.079188680648803,
      "epoch": 2.0205479452054793,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004589514402751103,
      "loss": 4.704,
      "mean_token_accuracy": 0.22483026534318923,
      "num_tokens": 44504021.0,
      "step": 24190
    },
    {
      "entropy": 5.058333969116211,
      "epoch": 2.0209655863681926,
      "grad_norm": 1.125,
      "learning_rate": 0.0004589342959019794,
      "loss": 4.7912,
      "mean_token_accuracy": 0.2148815542459488,
      "num_tokens": 44513337.0,
      "step": 24195
    },
    {
      "entropy": 5.028226089477539,
      "epoch": 2.0213832275309054,
      "grad_norm": 1.125,
      "learning_rate": 0.0004589171483088572,
      "loss": 4.7064,
      "mean_token_accuracy": 0.2281937450170517,
      "num_tokens": 44522206.0,
      "step": 24200
    },
    {
      "entropy": 5.075119972229004,
      "epoch": 2.0218008686936186,
      "grad_norm": 1.171875,
      "learning_rate": 0.000458899997496044,
      "loss": 4.8895,
      "mean_token_accuracy": 0.21078864187002183,
      "num_tokens": 44531030.0,
      "step": 24205
    },
    {
      "entropy": 5.164805269241333,
      "epoch": 2.0222185098563314,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045888284346384003,
      "loss": 4.8211,
      "mean_token_accuracy": 0.222714464366436,
      "num_tokens": 44539844.0,
      "step": 24210
    },
    {
      "entropy": 5.09724178314209,
      "epoch": 2.0226361510190443,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004588656862125454,
      "loss": 4.7057,
      "mean_token_accuracy": 0.2283868357539177,
      "num_tokens": 44548420.0,
      "step": 24215
    },
    {
      "entropy": 5.057922506332398,
      "epoch": 2.0230537921817575,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004588485257424608,
      "loss": 4.756,
      "mean_token_accuracy": 0.22811342030763626,
      "num_tokens": 44557597.0,
      "step": 24220
    },
    {
      "entropy": 5.028529310226441,
      "epoch": 2.0234714333444703,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004588313620538863,
      "loss": 4.7182,
      "mean_token_accuracy": 0.2246102586388588,
      "num_tokens": 44566281.0,
      "step": 24225
    },
    {
      "entropy": 5.092097902297974,
      "epoch": 2.0238890745071836,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004588141951471226,
      "loss": 4.8446,
      "mean_token_accuracy": 0.2126276522874832,
      "num_tokens": 44575937.0,
      "step": 24230
    },
    {
      "entropy": 5.119278764724731,
      "epoch": 2.0243067156698964,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045879702502247014,
      "loss": 4.8726,
      "mean_token_accuracy": 0.22601087093353273,
      "num_tokens": 44585373.0,
      "step": 24235
    },
    {
      "entropy": 5.146048974990845,
      "epoch": 2.024724356832609,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004587798516802296,
      "loss": 4.7994,
      "mean_token_accuracy": 0.21891558468341826,
      "num_tokens": 44594526.0,
      "step": 24240
    },
    {
      "entropy": 5.084617567062378,
      "epoch": 2.0251419979953225,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045876267512070146,
      "loss": 4.7956,
      "mean_token_accuracy": 0.22353945225477218,
      "num_tokens": 44603128.0,
      "step": 24245
    },
    {
      "entropy": 5.098733615875244,
      "epoch": 2.0255596391580353,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045874549534418644,
      "loss": 4.7639,
      "mean_token_accuracy": 0.2184221550822258,
      "num_tokens": 44612492.0,
      "step": 24250
    },
    {
      "entropy": 5.068977355957031,
      "epoch": 2.0259772803207485,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004587283123509855,
      "loss": 4.7751,
      "mean_token_accuracy": 0.21086640805006027,
      "num_tokens": 44621086.0,
      "step": 24255
    },
    {
      "entropy": 5.004705858230591,
      "epoch": 2.0263949214834613,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004587111261413992,
      "loss": 4.6971,
      "mean_token_accuracy": 0.2223461776971817,
      "num_tokens": 44629813.0,
      "step": 24260
    },
    {
      "entropy": 5.109647035598755,
      "epoch": 2.0268125626461746,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004586939367157285,
      "loss": 4.8799,
      "mean_token_accuracy": 0.2105504646897316,
      "num_tokens": 44638888.0,
      "step": 24265
    },
    {
      "entropy": 5.075044584274292,
      "epoch": 2.0272302038088874,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00045867674407427435,
      "loss": 4.7057,
      "mean_token_accuracy": 0.22883529663085939,
      "num_tokens": 44647617.0,
      "step": 24270
    },
    {
      "entropy": 5.11276388168335,
      "epoch": 2.0276478449716,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045865954821733767,
      "loss": 4.707,
      "mean_token_accuracy": 0.22197289317846297,
      "num_tokens": 44657456.0,
      "step": 24275
    },
    {
      "entropy": 5.058644008636475,
      "epoch": 2.0280654861343135,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045864234914521956,
      "loss": 4.7473,
      "mean_token_accuracy": 0.2192407637834549,
      "num_tokens": 44667008.0,
      "step": 24280
    },
    {
      "entropy": 5.041268348693848,
      "epoch": 2.0284831272970263,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004586251468582211,
      "loss": 4.7561,
      "mean_token_accuracy": 0.22196945548057556,
      "num_tokens": 44675448.0,
      "step": 24285
    },
    {
      "entropy": 5.123454427719116,
      "epoch": 2.0289007684597395,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004586079413566434,
      "loss": 4.8568,
      "mean_token_accuracy": 0.21452341824769974,
      "num_tokens": 44684540.0,
      "step": 24290
    },
    {
      "entropy": 5.188871955871582,
      "epoch": 2.0293184096224524,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045859073264078765,
      "loss": 4.8283,
      "mean_token_accuracy": 0.2142779067158699,
      "num_tokens": 44693734.0,
      "step": 24295
    },
    {
      "entropy": 5.10165843963623,
      "epoch": 2.029736050785165,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045857352071095514,
      "loss": 4.7614,
      "mean_token_accuracy": 0.22136759310960769,
      "num_tokens": 44702407.0,
      "step": 24300
    },
    {
      "entropy": 5.039311313629151,
      "epoch": 2.0301536919478784,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004585563055674472,
      "loss": 4.6681,
      "mean_token_accuracy": 0.2267340525984764,
      "num_tokens": 44711511.0,
      "step": 24305
    },
    {
      "entropy": 5.090967226028442,
      "epoch": 2.0305713331105912,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045853908721056526,
      "loss": 4.7797,
      "mean_token_accuracy": 0.21659083366394044,
      "num_tokens": 44720069.0,
      "step": 24310
    },
    {
      "entropy": 5.007187032699585,
      "epoch": 2.0309889742733045,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045852186564061055,
      "loss": 4.6903,
      "mean_token_accuracy": 0.226674622297287,
      "num_tokens": 44728549.0,
      "step": 24315
    },
    {
      "entropy": 5.069675636291504,
      "epoch": 2.0314066154360173,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004585046408578847,
      "loss": 4.7862,
      "mean_token_accuracy": 0.21516508758068084,
      "num_tokens": 44737753.0,
      "step": 24320
    },
    {
      "entropy": 5.112228298187256,
      "epoch": 2.0318242565987306,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045848741286268923,
      "loss": 4.7506,
      "mean_token_accuracy": 0.2182184189558029,
      "num_tokens": 44747591.0,
      "step": 24325
    },
    {
      "entropy": 5.050670862197876,
      "epoch": 2.0322418977614434,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045847018165532576,
      "loss": 4.8359,
      "mean_token_accuracy": 0.2205819457769394,
      "num_tokens": 44756942.0,
      "step": 24330
    },
    {
      "entropy": 5.036666440963745,
      "epoch": 2.032659538924156,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045845294723609596,
      "loss": 4.7639,
      "mean_token_accuracy": 0.21831498742103578,
      "num_tokens": 44765492.0,
      "step": 24335
    },
    {
      "entropy": 5.071583890914917,
      "epoch": 2.0330771800868694,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004584357096053014,
      "loss": 4.6775,
      "mean_token_accuracy": 0.22266904413700103,
      "num_tokens": 44774433.0,
      "step": 24340
    },
    {
      "entropy": 5.1731359481811525,
      "epoch": 2.0334948212495823,
      "grad_norm": 1.109375,
      "learning_rate": 0.000458418468763244,
      "loss": 4.8604,
      "mean_token_accuracy": 0.2154455929994583,
      "num_tokens": 44784107.0,
      "step": 24345
    },
    {
      "entropy": 5.088107681274414,
      "epoch": 2.0339124624122955,
      "grad_norm": 1.0,
      "learning_rate": 0.0004584012247102255,
      "loss": 4.7673,
      "mean_token_accuracy": 0.21874666959047318,
      "num_tokens": 44794234.0,
      "step": 24350
    },
    {
      "entropy": 4.980257844924926,
      "epoch": 2.0343301035750083,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004583839774465479,
      "loss": 4.6621,
      "mean_token_accuracy": 0.22496493756771088,
      "num_tokens": 44802881.0,
      "step": 24355
    },
    {
      "entropy": 4.997206830978394,
      "epoch": 2.034747744737721,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045836672697251293,
      "loss": 4.7423,
      "mean_token_accuracy": 0.22435784190893174,
      "num_tokens": 44812114.0,
      "step": 24360
    },
    {
      "entropy": 5.086582612991333,
      "epoch": 2.0351653859004344,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004583494732884227,
      "loss": 4.8083,
      "mean_token_accuracy": 0.21602927297353744,
      "num_tokens": 44821227.0,
      "step": 24365
    },
    {
      "entropy": 5.135420894622802,
      "epoch": 2.035583027063147,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004583322163945792,
      "loss": 4.7407,
      "mean_token_accuracy": 0.2225087359547615,
      "num_tokens": 44829703.0,
      "step": 24370
    },
    {
      "entropy": 5.091836643218994,
      "epoch": 2.0360006682258605,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004583149562912847,
      "loss": 4.8203,
      "mean_token_accuracy": 0.21781303882598876,
      "num_tokens": 44839400.0,
      "step": 24375
    },
    {
      "entropy": 4.993625640869141,
      "epoch": 2.0364183093885733,
      "grad_norm": 1.25,
      "learning_rate": 0.0004582976929788411,
      "loss": 4.728,
      "mean_token_accuracy": 0.22282288670539857,
      "num_tokens": 44848994.0,
      "step": 24380
    },
    {
      "entropy": 5.073931550979614,
      "epoch": 2.0368359505512865,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004582804264575508,
      "loss": 4.7364,
      "mean_token_accuracy": 0.2218945726752281,
      "num_tokens": 44857562.0,
      "step": 24385
    },
    {
      "entropy": 5.146783447265625,
      "epoch": 2.0372535917139993,
      "grad_norm": 1.125,
      "learning_rate": 0.000458263156727716,
      "loss": 4.8411,
      "mean_token_accuracy": 0.21933116614818574,
      "num_tokens": 44867597.0,
      "step": 24390
    },
    {
      "entropy": 5.05325288772583,
      "epoch": 2.037671232876712,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004582458837896391,
      "loss": 4.7826,
      "mean_token_accuracy": 0.21708951145410538,
      "num_tokens": 44876678.0,
      "step": 24395
    },
    {
      "entropy": 5.011784458160401,
      "epoch": 2.0380888740394254,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045822860764362235,
      "loss": 4.7054,
      "mean_token_accuracy": 0.22505275309085845,
      "num_tokens": 44885847.0,
      "step": 24400
    },
    {
      "entropy": 5.111807775497437,
      "epoch": 2.038506515202138,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004582113282899683,
      "loss": 4.8186,
      "mean_token_accuracy": 0.21542145162820817,
      "num_tokens": 44896078.0,
      "step": 24405
    },
    {
      "entropy": 5.034661293029785,
      "epoch": 2.0389241563648515,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004581940457289795,
      "loss": 4.6442,
      "mean_token_accuracy": 0.23154675513505935,
      "num_tokens": 44904996.0,
      "step": 24410
    },
    {
      "entropy": 4.993531084060669,
      "epoch": 2.0393417975275643,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004581767599609584,
      "loss": 4.6868,
      "mean_token_accuracy": 0.2298189401626587,
      "num_tokens": 44913488.0,
      "step": 24415
    },
    {
      "entropy": 5.142660093307495,
      "epoch": 2.039759438690277,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004581594709862076,
      "loss": 4.9027,
      "mean_token_accuracy": 0.20600817650556563,
      "num_tokens": 44922836.0,
      "step": 24420
    },
    {
      "entropy": 5.029830646514893,
      "epoch": 2.0401770798529903,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045814217880502973,
      "loss": 4.683,
      "mean_token_accuracy": 0.22428359240293502,
      "num_tokens": 44932494.0,
      "step": 24425
    },
    {
      "entropy": 5.059798526763916,
      "epoch": 2.040594721015703,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004581248834177276,
      "loss": 4.7087,
      "mean_token_accuracy": 0.2213624343276024,
      "num_tokens": 44941572.0,
      "step": 24430
    },
    {
      "entropy": 5.02242751121521,
      "epoch": 2.0410123621784164,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045810758482460397,
      "loss": 4.7418,
      "mean_token_accuracy": 0.22033767849206926,
      "num_tokens": 44949914.0,
      "step": 24435
    },
    {
      "entropy": 5.073132705688477,
      "epoch": 2.0414300033411292,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004580902830259617,
      "loss": 4.7861,
      "mean_token_accuracy": 0.21885644346475602,
      "num_tokens": 44958726.0,
      "step": 24440
    },
    {
      "entropy": 5.132407236099243,
      "epoch": 2.0418476445038425,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045807297802210364,
      "loss": 4.8119,
      "mean_token_accuracy": 0.2232871398329735,
      "num_tokens": 44967204.0,
      "step": 24445
    },
    {
      "entropy": 5.026123523712158,
      "epoch": 2.0422652856665553,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004580556698133327,
      "loss": 4.7682,
      "mean_token_accuracy": 0.21399967968463898,
      "num_tokens": 44977043.0,
      "step": 24450
    },
    {
      "entropy": 5.1292649269104,
      "epoch": 2.042682926829268,
      "grad_norm": 1.25,
      "learning_rate": 0.0004580383583999519,
      "loss": 4.8258,
      "mean_token_accuracy": 0.20722326040267944,
      "num_tokens": 44986762.0,
      "step": 24455
    },
    {
      "entropy": 5.103384208679199,
      "epoch": 2.0431005679919814,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004580210437822643,
      "loss": 4.753,
      "mean_token_accuracy": 0.2160036027431488,
      "num_tokens": 44995513.0,
      "step": 24460
    },
    {
      "entropy": 5.087836742401123,
      "epoch": 2.043518209154694,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004580037259605731,
      "loss": 4.8876,
      "mean_token_accuracy": 0.21891800910234452,
      "num_tokens": 45005127.0,
      "step": 24465
    },
    {
      "entropy": 5.1455872535705565,
      "epoch": 2.0439358503174074,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004579864049351813,
      "loss": 4.7784,
      "mean_token_accuracy": 0.22063252776861192,
      "num_tokens": 45013994.0,
      "step": 24470
    },
    {
      "entropy": 5.0517919063568115,
      "epoch": 2.0443534914801202,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004579690807063923,
      "loss": 4.8165,
      "mean_token_accuracy": 0.21535662710666656,
      "num_tokens": 45024177.0,
      "step": 24475
    },
    {
      "entropy": 5.102361583709717,
      "epoch": 2.0447711326428335,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004579517532745093,
      "loss": 4.7853,
      "mean_token_accuracy": 0.21532706022262574,
      "num_tokens": 45033139.0,
      "step": 24480
    },
    {
      "entropy": 5.150422430038452,
      "epoch": 2.0451887738055463,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045793442263983563,
      "loss": 4.8571,
      "mean_token_accuracy": 0.21223242729902267,
      "num_tokens": 45043394.0,
      "step": 24485
    },
    {
      "entropy": 5.110674142837524,
      "epoch": 2.045606414968259,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004579170888026747,
      "loss": 4.7614,
      "mean_token_accuracy": 0.2193707600235939,
      "num_tokens": 45052015.0,
      "step": 24490
    },
    {
      "entropy": 4.99463906288147,
      "epoch": 2.0460240561309724,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004578997517633299,
      "loss": 4.6803,
      "mean_token_accuracy": 0.22125990092754363,
      "num_tokens": 45060836.0,
      "step": 24495
    },
    {
      "entropy": 5.0331755638122555,
      "epoch": 2.046441697293685,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045788241152210486,
      "loss": 4.7239,
      "mean_token_accuracy": 0.21735795885324477,
      "num_tokens": 45070007.0,
      "step": 24500
    },
    {
      "entropy": 5.153605556488037,
      "epoch": 2.0468593384563984,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000457865068079303,
      "loss": 4.823,
      "mean_token_accuracy": 0.21833566278219224,
      "num_tokens": 45079984.0,
      "step": 24505
    },
    {
      "entropy": 5.013723564147949,
      "epoch": 2.0472769796191113,
      "grad_norm": 1.328125,
      "learning_rate": 0.000457847721435228,
      "loss": 4.6965,
      "mean_token_accuracy": 0.22280552089214326,
      "num_tokens": 45088676.0,
      "step": 24510
    },
    {
      "entropy": 5.122201156616211,
      "epoch": 2.047694620781824,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045783037159018364,
      "loss": 4.834,
      "mean_token_accuracy": 0.21764011085033416,
      "num_tokens": 45098232.0,
      "step": 24515
    },
    {
      "entropy": 5.0511061668396,
      "epoch": 2.0481122619445373,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045781301854447347,
      "loss": 4.7711,
      "mean_token_accuracy": 0.22463646531105042,
      "num_tokens": 45107789.0,
      "step": 24520
    },
    {
      "entropy": 5.085722494125366,
      "epoch": 2.04852990310725,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045779566229840144,
      "loss": 4.7694,
      "mean_token_accuracy": 0.2165216788649559,
      "num_tokens": 45117824.0,
      "step": 24525
    },
    {
      "entropy": 5.19319109916687,
      "epoch": 2.0489475442699634,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045777830285227125,
      "loss": 4.9084,
      "mean_token_accuracy": 0.20899424105882644,
      "num_tokens": 45127625.0,
      "step": 24530
    },
    {
      "entropy": 5.054113864898682,
      "epoch": 2.049365185432676,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004577609402063868,
      "loss": 4.7178,
      "mean_token_accuracy": 0.22478894144296646,
      "num_tokens": 45135931.0,
      "step": 24535
    },
    {
      "entropy": 5.114055013656616,
      "epoch": 2.049782826595389,
      "grad_norm": 1.25,
      "learning_rate": 0.0004577435743610522,
      "loss": 4.8556,
      "mean_token_accuracy": 0.21751478016376496,
      "num_tokens": 45145690.0,
      "step": 24540
    },
    {
      "entropy": 5.155566549301147,
      "epoch": 2.0502004677581023,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004577262053165713,
      "loss": 4.942,
      "mean_token_accuracy": 0.20636038482189178,
      "num_tokens": 45155455.0,
      "step": 24545
    },
    {
      "entropy": 5.1255943298339846,
      "epoch": 2.050618108920815,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004577088330732483,
      "loss": 4.7993,
      "mean_token_accuracy": 0.22505345195531845,
      "num_tokens": 45164377.0,
      "step": 24550
    },
    {
      "entropy": 5.201976013183594,
      "epoch": 2.0510357500835283,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004576914576313872,
      "loss": 4.8398,
      "mean_token_accuracy": 0.21174152046442032,
      "num_tokens": 45173037.0,
      "step": 24555
    },
    {
      "entropy": 5.116738986968994,
      "epoch": 2.051453391246241,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004576740789912922,
      "loss": 4.7666,
      "mean_token_accuracy": 0.22163022309541702,
      "num_tokens": 45182459.0,
      "step": 24560
    },
    {
      "entropy": 5.021659708023071,
      "epoch": 2.0518710324089544,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004576566971532676,
      "loss": 4.6878,
      "mean_token_accuracy": 0.2296021595597267,
      "num_tokens": 45190744.0,
      "step": 24565
    },
    {
      "entropy": 5.098752212524414,
      "epoch": 2.052288673571667,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004576393121176177,
      "loss": 4.8889,
      "mean_token_accuracy": 0.2132757395505905,
      "num_tokens": 45198852.0,
      "step": 24570
    },
    {
      "entropy": 5.039467000961304,
      "epoch": 2.05270631473438,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045762192388464675,
      "loss": 4.7061,
      "mean_token_accuracy": 0.22921119183301925,
      "num_tokens": 45207794.0,
      "step": 24575
    },
    {
      "entropy": 5.046278667449951,
      "epoch": 2.0531239558970933,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004576045324546592,
      "loss": 4.7227,
      "mean_token_accuracy": 0.23017247766256332,
      "num_tokens": 45216621.0,
      "step": 24580
    },
    {
      "entropy": 5.08564944267273,
      "epoch": 2.053541597059806,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045758713782795946,
      "loss": 4.8369,
      "mean_token_accuracy": 0.21343822479248048,
      "num_tokens": 45226801.0,
      "step": 24585
    },
    {
      "entropy": 5.047116231918335,
      "epoch": 2.0539592382225194,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045756974000485216,
      "loss": 4.7797,
      "mean_token_accuracy": 0.2102496400475502,
      "num_tokens": 45236575.0,
      "step": 24590
    },
    {
      "entropy": 5.114897680282593,
      "epoch": 2.054376879385232,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045755233898564176,
      "loss": 4.813,
      "mean_token_accuracy": 0.21279629170894623,
      "num_tokens": 45245868.0,
      "step": 24595
    },
    {
      "entropy": 5.149668169021607,
      "epoch": 2.0547945205479454,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045753493477063297,
      "loss": 4.8606,
      "mean_token_accuracy": 0.21236229836940765,
      "num_tokens": 45256160.0,
      "step": 24600
    },
    {
      "entropy": 5.0374058246612545,
      "epoch": 2.0552121617106582,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004575175273601304,
      "loss": 4.7579,
      "mean_token_accuracy": 0.22103066444396974,
      "num_tokens": 45265128.0,
      "step": 24605
    },
    {
      "entropy": 4.983345651626587,
      "epoch": 2.055629802873371,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045750011675443886,
      "loss": 4.7115,
      "mean_token_accuracy": 0.2194851189851761,
      "num_tokens": 45274825.0,
      "step": 24610
    },
    {
      "entropy": 5.008757638931274,
      "epoch": 2.0560474440360843,
      "grad_norm": 1.25,
      "learning_rate": 0.0004574827029538631,
      "loss": 4.6832,
      "mean_token_accuracy": 0.23008553683757782,
      "num_tokens": 45282585.0,
      "step": 24615
    },
    {
      "entropy": 5.063090658187866,
      "epoch": 2.056465085198797,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004574652859587079,
      "loss": 4.7999,
      "mean_token_accuracy": 0.2173866704106331,
      "num_tokens": 45292752.0,
      "step": 24620
    },
    {
      "entropy": 5.085232067108154,
      "epoch": 2.0568827263615104,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045744786576927825,
      "loss": 4.6908,
      "mean_token_accuracy": 0.22865095734596252,
      "num_tokens": 45302486.0,
      "step": 24625
    },
    {
      "entropy": 5.063835382461548,
      "epoch": 2.057300367524223,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004574304423858791,
      "loss": 4.8255,
      "mean_token_accuracy": 0.2159842848777771,
      "num_tokens": 45311215.0,
      "step": 24630
    },
    {
      "entropy": 5.0670593738555905,
      "epoch": 2.057718008686936,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045741301580881557,
      "loss": 4.7954,
      "mean_token_accuracy": 0.2150513559579849,
      "num_tokens": 45321800.0,
      "step": 24635
    },
    {
      "entropy": 5.002824354171753,
      "epoch": 2.0581356498496493,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004573955860383925,
      "loss": 4.752,
      "mean_token_accuracy": 0.2283237621188164,
      "num_tokens": 45330621.0,
      "step": 24640
    },
    {
      "entropy": 5.0763946056365965,
      "epoch": 2.058553291012362,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004573781530749152,
      "loss": 4.7444,
      "mean_token_accuracy": 0.22343235611915588,
      "num_tokens": 45340295.0,
      "step": 24645
    },
    {
      "entropy": 5.075274705886841,
      "epoch": 2.0589709321750753,
      "grad_norm": 1.125,
      "learning_rate": 0.00045736071691868883,
      "loss": 4.8277,
      "mean_token_accuracy": 0.215005961060524,
      "num_tokens": 45350896.0,
      "step": 24650
    },
    {
      "entropy": 5.009995222091675,
      "epoch": 2.059388573337788,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004573432775700185,
      "loss": 4.6694,
      "mean_token_accuracy": 0.23028864860534667,
      "num_tokens": 45359462.0,
      "step": 24655
    },
    {
      "entropy": 5.1729615211486815,
      "epoch": 2.0598062145005014,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004573258350292097,
      "loss": 4.8391,
      "mean_token_accuracy": 0.2089538037776947,
      "num_tokens": 45370272.0,
      "step": 24660
    },
    {
      "entropy": 5.093541097640991,
      "epoch": 2.060223855663214,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004573083892965677,
      "loss": 4.7851,
      "mean_token_accuracy": 0.22777036726474761,
      "num_tokens": 45379745.0,
      "step": 24665
    },
    {
      "entropy": 4.98001012802124,
      "epoch": 2.060641496825927,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004572909403723978,
      "loss": 4.6215,
      "mean_token_accuracy": 0.23727313727140426,
      "num_tokens": 45388599.0,
      "step": 24670
    },
    {
      "entropy": 5.014883804321289,
      "epoch": 2.0610591379886403,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004572734882570057,
      "loss": 4.7223,
      "mean_token_accuracy": 0.21819277852773666,
      "num_tokens": 45397246.0,
      "step": 24675
    },
    {
      "entropy": 5.125032186508179,
      "epoch": 2.061476779151353,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045725603295069666,
      "loss": 4.8152,
      "mean_token_accuracy": 0.20783943682909012,
      "num_tokens": 45406001.0,
      "step": 24680
    },
    {
      "entropy": 4.985892915725708,
      "epoch": 2.0618944203140663,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004572385744537765,
      "loss": 4.7086,
      "mean_token_accuracy": 0.22175046801567078,
      "num_tokens": 45414926.0,
      "step": 24685
    },
    {
      "entropy": 5.115460443496704,
      "epoch": 2.062312061476779,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00045722111276655066,
      "loss": 4.792,
      "mean_token_accuracy": 0.21520153880119325,
      "num_tokens": 45423918.0,
      "step": 24690
    },
    {
      "entropy": 4.932935571670532,
      "epoch": 2.062729702639492,
      "grad_norm": 1.125,
      "learning_rate": 0.00045720364788932486,
      "loss": 4.647,
      "mean_token_accuracy": 0.22569793164730073,
      "num_tokens": 45433174.0,
      "step": 24695
    },
    {
      "entropy": 4.995977830886841,
      "epoch": 2.063147343802205,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045718617982240496,
      "loss": 4.7502,
      "mean_token_accuracy": 0.22295697182416915,
      "num_tokens": 45441660.0,
      "step": 24700
    },
    {
      "entropy": 5.138670158386231,
      "epoch": 2.063564984964918,
      "grad_norm": 1.125,
      "learning_rate": 0.0004571687085660967,
      "loss": 4.7871,
      "mean_token_accuracy": 0.21412858217954636,
      "num_tokens": 45451293.0,
      "step": 24705
    },
    {
      "entropy": 5.129530525207519,
      "epoch": 2.0639826261276313,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045715123412070594,
      "loss": 4.7591,
      "mean_token_accuracy": 0.2157309353351593,
      "num_tokens": 45460615.0,
      "step": 24710
    },
    {
      "entropy": 5.10470552444458,
      "epoch": 2.064400267290344,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004571337564865385,
      "loss": 4.7613,
      "mean_token_accuracy": 0.21844542324542998,
      "num_tokens": 45469404.0,
      "step": 24715
    },
    {
      "entropy": 5.160469198226929,
      "epoch": 2.0648179084530573,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004571162756639004,
      "loss": 4.8389,
      "mean_token_accuracy": 0.22176022529602052,
      "num_tokens": 45478900.0,
      "step": 24720
    },
    {
      "entropy": 4.981995296478272,
      "epoch": 2.06523554961577,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004570987916530977,
      "loss": 4.6655,
      "mean_token_accuracy": 0.2191343918442726,
      "num_tokens": 45488478.0,
      "step": 24725
    },
    {
      "entropy": 5.02810435295105,
      "epoch": 2.065653190778483,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045708130445443646,
      "loss": 4.7105,
      "mean_token_accuracy": 0.21393455117940902,
      "num_tokens": 45498421.0,
      "step": 24730
    },
    {
      "entropy": 5.064799165725708,
      "epoch": 2.0660708319411962,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045706381406822284,
      "loss": 4.7555,
      "mean_token_accuracy": 0.21406081765890123,
      "num_tokens": 45507715.0,
      "step": 24735
    },
    {
      "entropy": 5.101679801940918,
      "epoch": 2.066488473103909,
      "grad_norm": 1.09375,
      "learning_rate": 0.000457046320494763,
      "loss": 4.7805,
      "mean_token_accuracy": 0.21853684037923812,
      "num_tokens": 45516524.0,
      "step": 24740
    },
    {
      "entropy": 5.073093318939209,
      "epoch": 2.0669061142666223,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00045702882373436317,
      "loss": 4.758,
      "mean_token_accuracy": 0.21853466033935548,
      "num_tokens": 45526058.0,
      "step": 24745
    },
    {
      "entropy": 5.06175684928894,
      "epoch": 2.067323755429335,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004570113237873297,
      "loss": 4.8084,
      "mean_token_accuracy": 0.21754117161035538,
      "num_tokens": 45535441.0,
      "step": 24750
    },
    {
      "entropy": 5.116645669937133,
      "epoch": 2.067741396592048,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045699382065396893,
      "loss": 4.8699,
      "mean_token_accuracy": 0.2133322149515152,
      "num_tokens": 45544355.0,
      "step": 24755
    },
    {
      "entropy": 5.099468803405761,
      "epoch": 2.068159037754761,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004569763143345872,
      "loss": 4.7371,
      "mean_token_accuracy": 0.2213921219110489,
      "num_tokens": 45553509.0,
      "step": 24760
    },
    {
      "entropy": 4.986692094802857,
      "epoch": 2.068576678917474,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045695880482949107,
      "loss": 4.6127,
      "mean_token_accuracy": 0.23044623732566832,
      "num_tokens": 45563016.0,
      "step": 24765
    },
    {
      "entropy": 5.039471578598023,
      "epoch": 2.0689943200801872,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004569412921389871,
      "loss": 4.771,
      "mean_token_accuracy": 0.21764363050460817,
      "num_tokens": 45572242.0,
      "step": 24770
    },
    {
      "entropy": 5.026662111282349,
      "epoch": 2.0694119612429,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004569237762633818,
      "loss": 4.6346,
      "mean_token_accuracy": 0.2228830114006996,
      "num_tokens": 45581430.0,
      "step": 24775
    },
    {
      "entropy": 5.056074571609497,
      "epoch": 2.0698296024056133,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045690625720298166,
      "loss": 4.7297,
      "mean_token_accuracy": 0.21872189939022063,
      "num_tokens": 45590161.0,
      "step": 24780
    },
    {
      "entropy": 5.04354772567749,
      "epoch": 2.070247243568326,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004568887349580937,
      "loss": 4.8171,
      "mean_token_accuracy": 0.21229731291532516,
      "num_tokens": 45599243.0,
      "step": 24785
    },
    {
      "entropy": 5.044265508651733,
      "epoch": 2.070664884731039,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045687120952902447,
      "loss": 4.7505,
      "mean_token_accuracy": 0.22098505049943923,
      "num_tokens": 45607958.0,
      "step": 24790
    },
    {
      "entropy": 5.131627702713013,
      "epoch": 2.071082525893752,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045685368091608075,
      "loss": 4.8383,
      "mean_token_accuracy": 0.2171132519841194,
      "num_tokens": 45617243.0,
      "step": 24795
    },
    {
      "entropy": 5.055289268493652,
      "epoch": 2.071500167056465,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045683614911956946,
      "loss": 4.7518,
      "mean_token_accuracy": 0.21759848296642303,
      "num_tokens": 45627288.0,
      "step": 24800
    },
    {
      "entropy": 5.068768835067749,
      "epoch": 2.0719178082191783,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004568186141397975,
      "loss": 4.8143,
      "mean_token_accuracy": 0.2172176033258438,
      "num_tokens": 45636345.0,
      "step": 24805
    },
    {
      "entropy": 5.0997497081756595,
      "epoch": 2.072335449381891,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045680107597707186,
      "loss": 4.7451,
      "mean_token_accuracy": 0.2240162342786789,
      "num_tokens": 45645146.0,
      "step": 24810
    },
    {
      "entropy": 5.110265779495239,
      "epoch": 2.072753090544604,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045678353463169953,
      "loss": 4.8407,
      "mean_token_accuracy": 0.21413324773311615,
      "num_tokens": 45654509.0,
      "step": 24815
    },
    {
      "entropy": 5.042325401306153,
      "epoch": 2.073170731707317,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004567659901039876,
      "loss": 4.7024,
      "mean_token_accuracy": 0.22455477565526963,
      "num_tokens": 45663021.0,
      "step": 24820
    },
    {
      "entropy": 5.033304691314697,
      "epoch": 2.07358837287003,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004567484423942433,
      "loss": 4.7744,
      "mean_token_accuracy": 0.22112215161323548,
      "num_tokens": 45671633.0,
      "step": 24825
    },
    {
      "entropy": 5.137353515625,
      "epoch": 2.074006014032743,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045673089150277365,
      "loss": 4.8681,
      "mean_token_accuracy": 0.21082197576761247,
      "num_tokens": 45681974.0,
      "step": 24830
    },
    {
      "entropy": 5.097488307952881,
      "epoch": 2.074423655195456,
      "grad_norm": 1.125,
      "learning_rate": 0.00045671333742988604,
      "loss": 4.7595,
      "mean_token_accuracy": 0.21448614597320556,
      "num_tokens": 45691939.0,
      "step": 24835
    },
    {
      "entropy": 5.27419867515564,
      "epoch": 2.0748412963581693,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004566957801758878,
      "loss": 4.9279,
      "mean_token_accuracy": 0.21125391572713853,
      "num_tokens": 45701689.0,
      "step": 24840
    },
    {
      "entropy": 4.974298334121704,
      "epoch": 2.075258937520882,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004566782197410861,
      "loss": 4.6171,
      "mean_token_accuracy": 0.23247787803411485,
      "num_tokens": 45710523.0,
      "step": 24845
    },
    {
      "entropy": 5.051559162139893,
      "epoch": 2.075676578683595,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004566606561257885,
      "loss": 4.7856,
      "mean_token_accuracy": 0.2182942196726799,
      "num_tokens": 45719422.0,
      "step": 24850
    },
    {
      "entropy": 5.084012508392334,
      "epoch": 2.076094219846308,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004566430893303025,
      "loss": 4.7936,
      "mean_token_accuracy": 0.21825681924819945,
      "num_tokens": 45729052.0,
      "step": 24855
    },
    {
      "entropy": 5.036136341094971,
      "epoch": 2.076511861009021,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004566255193549355,
      "loss": 4.7898,
      "mean_token_accuracy": 0.2164666935801506,
      "num_tokens": 45739853.0,
      "step": 24860
    },
    {
      "entropy": 5.008579874038697,
      "epoch": 2.076929502171734,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045660794619999524,
      "loss": 4.6579,
      "mean_token_accuracy": 0.23323535472154616,
      "num_tokens": 45749433.0,
      "step": 24865
    },
    {
      "entropy": 5.072321462631225,
      "epoch": 2.077347143334447,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004565903698657893,
      "loss": 4.7508,
      "mean_token_accuracy": 0.2170883297920227,
      "num_tokens": 45758771.0,
      "step": 24870
    },
    {
      "entropy": 5.10025839805603,
      "epoch": 2.07776478449716,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045657279035262534,
      "loss": 4.9004,
      "mean_token_accuracy": 0.20258543342351915,
      "num_tokens": 45767998.0,
      "step": 24875
    },
    {
      "entropy": 5.132742929458618,
      "epoch": 2.078182425659873,
      "grad_norm": 1.359375,
      "learning_rate": 0.000456555207660811,
      "loss": 4.8878,
      "mean_token_accuracy": 0.2228178158402443,
      "num_tokens": 45777482.0,
      "step": 24880
    },
    {
      "entropy": 5.154902076721191,
      "epoch": 2.078600066822586,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004565376217906543,
      "loss": 4.7809,
      "mean_token_accuracy": 0.2121901661157608,
      "num_tokens": 45786996.0,
      "step": 24885
    },
    {
      "entropy": 5.179261350631714,
      "epoch": 2.079017707985299,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045652003274246306,
      "loss": 4.8996,
      "mean_token_accuracy": 0.21204865574836732,
      "num_tokens": 45796654.0,
      "step": 24890
    },
    {
      "entropy": 5.0684082984924315,
      "epoch": 2.079435349148012,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004565024405165452,
      "loss": 4.7791,
      "mean_token_accuracy": 0.22245754152536393,
      "num_tokens": 45805086.0,
      "step": 24895
    },
    {
      "entropy": 5.09503436088562,
      "epoch": 2.0798529903107252,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045648484511320853,
      "loss": 4.7749,
      "mean_token_accuracy": 0.22253199368715287,
      "num_tokens": 45813034.0,
      "step": 24900
    },
    {
      "entropy": 4.97293643951416,
      "epoch": 2.080270631473438,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045646724653276117,
      "loss": 4.6567,
      "mean_token_accuracy": 0.22567058503627777,
      "num_tokens": 45821867.0,
      "step": 24905
    },
    {
      "entropy": 4.966631507873535,
      "epoch": 2.080688272636151,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045644964477551124,
      "loss": 4.6189,
      "mean_token_accuracy": 0.23236018270254136,
      "num_tokens": 45830640.0,
      "step": 24910
    },
    {
      "entropy": 5.014826107025146,
      "epoch": 2.081105913798864,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000456432039841767,
      "loss": 4.6889,
      "mean_token_accuracy": 0.22365398406982423,
      "num_tokens": 45839334.0,
      "step": 24915
    },
    {
      "entropy": 5.1258689880371096,
      "epoch": 2.081523554961577,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004564144317318364,
      "loss": 4.8146,
      "mean_token_accuracy": 0.2159166678786278,
      "num_tokens": 45849253.0,
      "step": 24920
    },
    {
      "entropy": 5.17208342552185,
      "epoch": 2.08194119612429,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045639682044602777,
      "loss": 4.7829,
      "mean_token_accuracy": 0.21279531419277192,
      "num_tokens": 45858166.0,
      "step": 24925
    },
    {
      "entropy": 5.117943239212036,
      "epoch": 2.082358837287003,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045637920598464956,
      "loss": 4.8845,
      "mean_token_accuracy": 0.21339764147996904,
      "num_tokens": 45867658.0,
      "step": 24930
    },
    {
      "entropy": 5.073947906494141,
      "epoch": 2.082776478449716,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045636158834800987,
      "loss": 4.7028,
      "mean_token_accuracy": 0.22497003823518752,
      "num_tokens": 45876817.0,
      "step": 24935
    },
    {
      "entropy": 5.075156307220459,
      "epoch": 2.083194119612429,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045634396753641736,
      "loss": 4.7487,
      "mean_token_accuracy": 0.2229306995868683,
      "num_tokens": 45886019.0,
      "step": 24940
    },
    {
      "entropy": 5.076701354980469,
      "epoch": 2.083611760775142,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004563263435501805,
      "loss": 4.8027,
      "mean_token_accuracy": 0.21633758246898652,
      "num_tokens": 45894917.0,
      "step": 24945
    },
    {
      "entropy": 5.0855566501617435,
      "epoch": 2.084029401937855,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004563087163896076,
      "loss": 4.6823,
      "mean_token_accuracy": 0.22704103142023085,
      "num_tokens": 45903727.0,
      "step": 24950
    },
    {
      "entropy": 4.969715023040772,
      "epoch": 2.084447043100568,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045629108605500747,
      "loss": 4.6828,
      "mean_token_accuracy": 0.22637709826231003,
      "num_tokens": 45912217.0,
      "step": 24955
    },
    {
      "entropy": 5.060718536376953,
      "epoch": 2.084864684263281,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045627345254668864,
      "loss": 4.8388,
      "mean_token_accuracy": 0.2145158037543297,
      "num_tokens": 45921526.0,
      "step": 24960
    },
    {
      "entropy": 5.098969984054565,
      "epoch": 2.085282325425994,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045625581586495977,
      "loss": 4.8586,
      "mean_token_accuracy": 0.21688072979450226,
      "num_tokens": 45931115.0,
      "step": 24965
    },
    {
      "entropy": 5.115434169769287,
      "epoch": 2.085699966588707,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004562381760101297,
      "loss": 4.8557,
      "mean_token_accuracy": 0.21455219984054566,
      "num_tokens": 45940323.0,
      "step": 24970
    },
    {
      "entropy": 5.102276182174682,
      "epoch": 2.08611760775142,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004562205329825072,
      "loss": 4.7551,
      "mean_token_accuracy": 0.21711453646421433,
      "num_tokens": 45949354.0,
      "step": 24975
    },
    {
      "entropy": 5.040036535263061,
      "epoch": 2.086535248914133,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004562028867824012,
      "loss": 4.7624,
      "mean_token_accuracy": 0.20974596589803696,
      "num_tokens": 45958899.0,
      "step": 24980
    },
    {
      "entropy": 5.072512578964234,
      "epoch": 2.086952890076846,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045618523741012044,
      "loss": 4.7408,
      "mean_token_accuracy": 0.2238178238272667,
      "num_tokens": 45967431.0,
      "step": 24985
    },
    {
      "entropy": 4.958353328704834,
      "epoch": 2.087370531239559,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004561675848659741,
      "loss": 4.7545,
      "mean_token_accuracy": 0.22547601759433747,
      "num_tokens": 45976185.0,
      "step": 24990
    },
    {
      "entropy": 4.9475555419921875,
      "epoch": 2.0877881724022718,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045614992915027103,
      "loss": 4.6008,
      "mean_token_accuracy": 0.23342891186475753,
      "num_tokens": 45984648.0,
      "step": 24995
    },
    {
      "entropy": 5.012949419021607,
      "epoch": 2.088205813564985,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045613227026332045,
      "loss": 4.6698,
      "mean_token_accuracy": 0.22394230514764785,
      "num_tokens": 45993724.0,
      "step": 25000
    },
    {
      "entropy": 5.035962200164795,
      "epoch": 2.088623454727698,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004561146082054314,
      "loss": 4.7257,
      "mean_token_accuracy": 0.21939070224761964,
      "num_tokens": 46002998.0,
      "step": 25005
    },
    {
      "entropy": 5.009625387191773,
      "epoch": 2.089041095890411,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004560969429769132,
      "loss": 4.7245,
      "mean_token_accuracy": 0.22605503499507903,
      "num_tokens": 46010843.0,
      "step": 25010
    },
    {
      "entropy": 5.021603059768677,
      "epoch": 2.089458737053124,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045607927457807503,
      "loss": 4.6573,
      "mean_token_accuracy": 0.22716492414474487,
      "num_tokens": 46019733.0,
      "step": 25015
    },
    {
      "entropy": 5.093092250823974,
      "epoch": 2.089876378215837,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004560616030092261,
      "loss": 4.8211,
      "mean_token_accuracy": 0.21840455532073974,
      "num_tokens": 46029754.0,
      "step": 25020
    },
    {
      "entropy": 5.01769790649414,
      "epoch": 2.09029401937855,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004560439282706759,
      "loss": 4.6606,
      "mean_token_accuracy": 0.22788120359182357,
      "num_tokens": 46037855.0,
      "step": 25025
    },
    {
      "entropy": 5.086738109588623,
      "epoch": 2.090711660541263,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045602625036273385,
      "loss": 4.7842,
      "mean_token_accuracy": 0.21337589174509047,
      "num_tokens": 46046176.0,
      "step": 25030
    },
    {
      "entropy": 5.158729982376099,
      "epoch": 2.091129301703976,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045600856928570944,
      "loss": 4.8501,
      "mean_token_accuracy": 0.22071715891361238,
      "num_tokens": 46055847.0,
      "step": 25035
    },
    {
      "entropy": 5.13087010383606,
      "epoch": 2.091546942866689,
      "grad_norm": 1.5,
      "learning_rate": 0.00045599088503991205,
      "loss": 4.7912,
      "mean_token_accuracy": 0.21520792245864867,
      "num_tokens": 46065763.0,
      "step": 25040
    },
    {
      "entropy": 5.075226306915283,
      "epoch": 2.091964584029402,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004559731976256514,
      "loss": 4.8147,
      "mean_token_accuracy": 0.2084924563765526,
      "num_tokens": 46074564.0,
      "step": 25045
    },
    {
      "entropy": 5.015893983840942,
      "epoch": 2.092382225192115,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045595550704323706,
      "loss": 4.7867,
      "mean_token_accuracy": 0.2230950266122818,
      "num_tokens": 46083627.0,
      "step": 25050
    },
    {
      "entropy": 5.021913242340088,
      "epoch": 2.0927998663548277,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045593781329297876,
      "loss": 4.8107,
      "mean_token_accuracy": 0.21490521878004074,
      "num_tokens": 46092997.0,
      "step": 25055
    },
    {
      "entropy": 5.177651119232178,
      "epoch": 2.093217507517541,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045592011637518625,
      "loss": 4.928,
      "mean_token_accuracy": 0.20788189470767976,
      "num_tokens": 46103266.0,
      "step": 25060
    },
    {
      "entropy": 5.0772068977355955,
      "epoch": 2.093635148680254,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045590241629016933,
      "loss": 4.63,
      "mean_token_accuracy": 0.2289051502943039,
      "num_tokens": 46112896.0,
      "step": 25065
    },
    {
      "entropy": 5.0101700782775875,
      "epoch": 2.094052789842967,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045588471303823787,
      "loss": 4.7349,
      "mean_token_accuracy": 0.23069108575582503,
      "num_tokens": 46121134.0,
      "step": 25070
    },
    {
      "entropy": 5.053437662124634,
      "epoch": 2.09447043100568,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045586700661970174,
      "loss": 4.752,
      "mean_token_accuracy": 0.22105701118707657,
      "num_tokens": 46130617.0,
      "step": 25075
    },
    {
      "entropy": 5.074312210083008,
      "epoch": 2.094888072168393,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045584929703487107,
      "loss": 4.795,
      "mean_token_accuracy": 0.2189629703760147,
      "num_tokens": 46140584.0,
      "step": 25080
    },
    {
      "entropy": 5.078569793701172,
      "epoch": 2.095305713331106,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004558315842840556,
      "loss": 4.6956,
      "mean_token_accuracy": 0.2224952220916748,
      "num_tokens": 46150039.0,
      "step": 25085
    },
    {
      "entropy": 5.080924081802368,
      "epoch": 2.0957233544938187,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004558138683675657,
      "loss": 4.7348,
      "mean_token_accuracy": 0.21795505434274673,
      "num_tokens": 46158899.0,
      "step": 25090
    },
    {
      "entropy": 5.0675324440002445,
      "epoch": 2.096140995656532,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045579614928571135,
      "loss": 4.7927,
      "mean_token_accuracy": 0.2128966435790062,
      "num_tokens": 46167813.0,
      "step": 25095
    },
    {
      "entropy": 4.999770975112915,
      "epoch": 2.096558636819245,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004557784270388028,
      "loss": 4.6819,
      "mean_token_accuracy": 0.22516243904829025,
      "num_tokens": 46176440.0,
      "step": 25100
    },
    {
      "entropy": 5.019576930999756,
      "epoch": 2.096976277981958,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045576070162715025,
      "loss": 4.6872,
      "mean_token_accuracy": 0.22621008157730102,
      "num_tokens": 46186354.0,
      "step": 25105
    },
    {
      "entropy": 5.050554466247559,
      "epoch": 2.097393919144671,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000455742973051064,
      "loss": 4.6999,
      "mean_token_accuracy": 0.2239138141274452,
      "num_tokens": 46194819.0,
      "step": 25110
    },
    {
      "entropy": 5.134314823150635,
      "epoch": 2.0978115603073837,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004557252413108545,
      "loss": 4.8819,
      "mean_token_accuracy": 0.20834551751613617,
      "num_tokens": 46205687.0,
      "step": 25115
    },
    {
      "entropy": 5.135827922821045,
      "epoch": 2.098229201470097,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045570750640683207,
      "loss": 4.8874,
      "mean_token_accuracy": 0.20749164521694183,
      "num_tokens": 46214990.0,
      "step": 25120
    },
    {
      "entropy": 5.01585283279419,
      "epoch": 2.0986468426328098,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004556897683393073,
      "loss": 4.7564,
      "mean_token_accuracy": 0.21471246480941772,
      "num_tokens": 46224325.0,
      "step": 25125
    },
    {
      "entropy": 5.091878414154053,
      "epoch": 2.099064483795523,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045567202710859056,
      "loss": 4.856,
      "mean_token_accuracy": 0.20898920446634292,
      "num_tokens": 46234199.0,
      "step": 25130
    },
    {
      "entropy": 5.146059179306031,
      "epoch": 2.099482124958236,
      "grad_norm": 1.125,
      "learning_rate": 0.0004556542827149926,
      "loss": 4.7425,
      "mean_token_accuracy": 0.22819896340370177,
      "num_tokens": 46243633.0,
      "step": 25135
    },
    {
      "entropy": 5.000447940826416,
      "epoch": 2.099899766120949,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004556365351588239,
      "loss": 4.633,
      "mean_token_accuracy": 0.2343410611152649,
      "num_tokens": 46252994.0,
      "step": 25140
    },
    {
      "entropy": 5.020513248443604,
      "epoch": 2.100317407283662,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045561878444039526,
      "loss": 4.8184,
      "mean_token_accuracy": 0.21810624450445176,
      "num_tokens": 46262455.0,
      "step": 25145
    },
    {
      "entropy": 5.099331378936768,
      "epoch": 2.1007350484463747,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045560103056001733,
      "loss": 4.8424,
      "mean_token_accuracy": 0.21298044174909592,
      "num_tokens": 46271864.0,
      "step": 25150
    },
    {
      "entropy": 5.103006410598755,
      "epoch": 2.101152689609088,
      "grad_norm": 1.125,
      "learning_rate": 0.000455583273518001,
      "loss": 4.7054,
      "mean_token_accuracy": 0.22513251900672912,
      "num_tokens": 46281244.0,
      "step": 25155
    },
    {
      "entropy": 5.0061116218566895,
      "epoch": 2.1015703307718008,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004555655133146571,
      "loss": 4.7004,
      "mean_token_accuracy": 0.22393887490034103,
      "num_tokens": 46291145.0,
      "step": 25160
    },
    {
      "entropy": 5.003392076492309,
      "epoch": 2.101987971934514,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004555477499502965,
      "loss": 4.7913,
      "mean_token_accuracy": 0.2219369128346443,
      "num_tokens": 46300484.0,
      "step": 25165
    },
    {
      "entropy": 5.103836441040039,
      "epoch": 2.102405613097227,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004555299834252303,
      "loss": 4.732,
      "mean_token_accuracy": 0.2280944600701332,
      "num_tokens": 46309598.0,
      "step": 25170
    },
    {
      "entropy": 5.174602842330932,
      "epoch": 2.1028232542599397,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004555122137397694,
      "loss": 4.8134,
      "mean_token_accuracy": 0.21228334307670593,
      "num_tokens": 46318410.0,
      "step": 25175
    },
    {
      "entropy": 5.167469835281372,
      "epoch": 2.103240895422653,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045549444089422486,
      "loss": 4.9367,
      "mean_token_accuracy": 0.20782221108675003,
      "num_tokens": 46327011.0,
      "step": 25180
    },
    {
      "entropy": 5.121884202957153,
      "epoch": 2.1036585365853657,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045547666488890794,
      "loss": 4.7976,
      "mean_token_accuracy": 0.21840576827526093,
      "num_tokens": 46335935.0,
      "step": 25185
    },
    {
      "entropy": 5.042918157577515,
      "epoch": 2.104076177748079,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004554588857241297,
      "loss": 4.7868,
      "mean_token_accuracy": 0.21963548958301543,
      "num_tokens": 46344820.0,
      "step": 25190
    },
    {
      "entropy": 5.023866033554077,
      "epoch": 2.104493818910792,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045544110340020155,
      "loss": 4.7079,
      "mean_token_accuracy": 0.22089714109897612,
      "num_tokens": 46353484.0,
      "step": 25195
    },
    {
      "entropy": 5.09674038887024,
      "epoch": 2.104911460073505,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004554233179174346,
      "loss": 4.7977,
      "mean_token_accuracy": 0.21041581481695176,
      "num_tokens": 46362539.0,
      "step": 25200
    },
    {
      "entropy": 5.094524097442627,
      "epoch": 2.105329101236218,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004554055292761403,
      "loss": 4.8089,
      "mean_token_accuracy": 0.2216820865869522,
      "num_tokens": 46372750.0,
      "step": 25205
    },
    {
      "entropy": 5.0897081851959225,
      "epoch": 2.1057467423989307,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004553877374766301,
      "loss": 4.7715,
      "mean_token_accuracy": 0.22186287939548494,
      "num_tokens": 46381891.0,
      "step": 25210
    },
    {
      "entropy": 5.0493896961212155,
      "epoch": 2.106164383561644,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045536994251921535,
      "loss": 4.7281,
      "mean_token_accuracy": 0.22279658019542695,
      "num_tokens": 46391013.0,
      "step": 25215
    },
    {
      "entropy": 5.056865310668945,
      "epoch": 2.1065820247243567,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004553521444042077,
      "loss": 4.7427,
      "mean_token_accuracy": 0.22052433341741562,
      "num_tokens": 46400527.0,
      "step": 25220
    },
    {
      "entropy": 5.098125791549682,
      "epoch": 2.10699966588707,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045533434313191866,
      "loss": 4.8529,
      "mean_token_accuracy": 0.21044470071792604,
      "num_tokens": 46409869.0,
      "step": 25225
    },
    {
      "entropy": 4.992417049407959,
      "epoch": 2.107417307049783,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045531653870265994,
      "loss": 4.6755,
      "mean_token_accuracy": 0.22854651361703873,
      "num_tokens": 46418644.0,
      "step": 25230
    },
    {
      "entropy": 5.010688018798828,
      "epoch": 2.1078349482124956,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004552987311167431,
      "loss": 4.7334,
      "mean_token_accuracy": 0.23215866833925247,
      "num_tokens": 46427475.0,
      "step": 25235
    },
    {
      "entropy": 5.197094869613648,
      "epoch": 2.108252589375209,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045528092037447994,
      "loss": 4.9203,
      "mean_token_accuracy": 0.21173668652772903,
      "num_tokens": 46437322.0,
      "step": 25240
    },
    {
      "entropy": 5.115324354171753,
      "epoch": 2.1086702305379217,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004552631064761822,
      "loss": 4.7453,
      "mean_token_accuracy": 0.22511185109615325,
      "num_tokens": 46446193.0,
      "step": 25245
    },
    {
      "entropy": 5.034795188903809,
      "epoch": 2.109087871700635,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004552452894221619,
      "loss": 4.6824,
      "mean_token_accuracy": 0.23046267479658128,
      "num_tokens": 46455351.0,
      "step": 25250
    },
    {
      "entropy": 5.042263650894165,
      "epoch": 2.1095055128633478,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045522746921273084,
      "loss": 4.7961,
      "mean_token_accuracy": 0.2126193091273308,
      "num_tokens": 46464056.0,
      "step": 25255
    },
    {
      "entropy": 5.02291522026062,
      "epoch": 2.109923154026061,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045520964584820097,
      "loss": 4.6752,
      "mean_token_accuracy": 0.22774039953947067,
      "num_tokens": 46472654.0,
      "step": 25260
    },
    {
      "entropy": 5.096627569198608,
      "epoch": 2.110340795188774,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004551918193288844,
      "loss": 4.8097,
      "mean_token_accuracy": 0.221335506439209,
      "num_tokens": 46482075.0,
      "step": 25265
    },
    {
      "entropy": 4.996083450317383,
      "epoch": 2.1107584363514866,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045517398965509297,
      "loss": 4.7248,
      "mean_token_accuracy": 0.22338307201862334,
      "num_tokens": 46490353.0,
      "step": 25270
    },
    {
      "entropy": 5.007040929794312,
      "epoch": 2.1111760775142,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045515615682713913,
      "loss": 4.6766,
      "mean_token_accuracy": 0.22710998654365538,
      "num_tokens": 46499332.0,
      "step": 25275
    },
    {
      "entropy": 5.077548265457153,
      "epoch": 2.1115937186769127,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045513832084533484,
      "loss": 4.8391,
      "mean_token_accuracy": 0.21441047936677932,
      "num_tokens": 46508824.0,
      "step": 25280
    },
    {
      "entropy": 5.1970751762390135,
      "epoch": 2.112011359839626,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004551204817099924,
      "loss": 4.8373,
      "mean_token_accuracy": 0.21439372152090072,
      "num_tokens": 46518262.0,
      "step": 25285
    },
    {
      "entropy": 5.091234874725342,
      "epoch": 2.1124290010023388,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045510263942142416,
      "loss": 4.8415,
      "mean_token_accuracy": 0.21605848371982575,
      "num_tokens": 46527463.0,
      "step": 25290
    },
    {
      "entropy": 5.080765151977539,
      "epoch": 2.1128466421650516,
      "grad_norm": 1.125,
      "learning_rate": 0.0004550847939799424,
      "loss": 4.7848,
      "mean_token_accuracy": 0.2182969942688942,
      "num_tokens": 46537380.0,
      "step": 25295
    },
    {
      "entropy": 5.105274963378906,
      "epoch": 2.113264283327765,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045506694538585956,
      "loss": 4.7419,
      "mean_token_accuracy": 0.22436818480491638,
      "num_tokens": 46546664.0,
      "step": 25300
    },
    {
      "entropy": 5.037616443634033,
      "epoch": 2.1136819244904776,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004550490936394881,
      "loss": 4.7631,
      "mean_token_accuracy": 0.2210922807455063,
      "num_tokens": 46555180.0,
      "step": 25305
    },
    {
      "entropy": 5.151648950576782,
      "epoch": 2.114099565653191,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004550312387411406,
      "loss": 4.9071,
      "mean_token_accuracy": 0.2085841104388237,
      "num_tokens": 46565450.0,
      "step": 25310
    },
    {
      "entropy": 5.137933588027954,
      "epoch": 2.1145172068159037,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004550133806911295,
      "loss": 4.8725,
      "mean_token_accuracy": 0.21610193252563475,
      "num_tokens": 46574453.0,
      "step": 25315
    },
    {
      "entropy": 5.130177974700928,
      "epoch": 2.114934847978617,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004549955194897674,
      "loss": 4.829,
      "mean_token_accuracy": 0.22220497876405715,
      "num_tokens": 46584341.0,
      "step": 25320
    },
    {
      "entropy": 5.091350936889649,
      "epoch": 2.11535248914133,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045497765513736717,
      "loss": 4.7353,
      "mean_token_accuracy": 0.21811312288045884,
      "num_tokens": 46593247.0,
      "step": 25325
    },
    {
      "entropy": 5.054707431793213,
      "epoch": 2.1157701303040426,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045495978763424147,
      "loss": 4.7849,
      "mean_token_accuracy": 0.2235860362648964,
      "num_tokens": 46602410.0,
      "step": 25330
    },
    {
      "entropy": 5.064453601837158,
      "epoch": 2.116187771466756,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045494191698070295,
      "loss": 4.7268,
      "mean_token_accuracy": 0.218549844622612,
      "num_tokens": 46611348.0,
      "step": 25335
    },
    {
      "entropy": 5.139907789230347,
      "epoch": 2.1166054126294687,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045492404317706473,
      "loss": 4.8367,
      "mean_token_accuracy": 0.21113614439964296,
      "num_tokens": 46621416.0,
      "step": 25340
    },
    {
      "entropy": 5.0902708053588865,
      "epoch": 2.117023053792182,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004549061662236394,
      "loss": 4.8319,
      "mean_token_accuracy": 0.20732705295085907,
      "num_tokens": 46631839.0,
      "step": 25345
    },
    {
      "entropy": 5.075817394256592,
      "epoch": 2.1174406949548947,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004548882861207402,
      "loss": 4.6907,
      "mean_token_accuracy": 0.236197729408741,
      "num_tokens": 46640662.0,
      "step": 25350
    },
    {
      "entropy": 5.041402959823609,
      "epoch": 2.1178583361176075,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045487040286868,
      "loss": 4.8082,
      "mean_token_accuracy": 0.21191989183425902,
      "num_tokens": 46649399.0,
      "step": 25355
    },
    {
      "entropy": 4.986692237854004,
      "epoch": 2.118275977280321,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045485251646777187,
      "loss": 4.6746,
      "mean_token_accuracy": 0.21812293976545333,
      "num_tokens": 46658020.0,
      "step": 25360
    },
    {
      "entropy": 5.045366477966309,
      "epoch": 2.1186936184430336,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004548346269183289,
      "loss": 4.7923,
      "mean_token_accuracy": 0.2172655627131462,
      "num_tokens": 46666815.0,
      "step": 25365
    },
    {
      "entropy": 5.088416051864624,
      "epoch": 2.119111259605747,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00045481673422066444,
      "loss": 4.7215,
      "mean_token_accuracy": 0.2230062872171402,
      "num_tokens": 46675582.0,
      "step": 25370
    },
    {
      "entropy": 5.145336580276489,
      "epoch": 2.1195289007684597,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004547988383750915,
      "loss": 4.9345,
      "mean_token_accuracy": 0.2122654214501381,
      "num_tokens": 46686385.0,
      "step": 25375
    },
    {
      "entropy": 5.145368528366089,
      "epoch": 2.119946541931173,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004547809393819235,
      "loss": 4.7925,
      "mean_token_accuracy": 0.21603647321462632,
      "num_tokens": 46695445.0,
      "step": 25380
    },
    {
      "entropy": 5.1735608100891115,
      "epoch": 2.1203641830938857,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004547630372414737,
      "loss": 4.7584,
      "mean_token_accuracy": 0.22007721960544585,
      "num_tokens": 46704562.0,
      "step": 25385
    },
    {
      "entropy": 5.077461862564087,
      "epoch": 2.1207818242565986,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004547451319540557,
      "loss": 4.814,
      "mean_token_accuracy": 0.21958725303411483,
      "num_tokens": 46713736.0,
      "step": 25390
    },
    {
      "entropy": 5.017127466201782,
      "epoch": 2.121199465419312,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004547272235199826,
      "loss": 4.7247,
      "mean_token_accuracy": 0.22137430906295777,
      "num_tokens": 46723160.0,
      "step": 25395
    },
    {
      "entropy": 5.049685621261597,
      "epoch": 2.1216171065820246,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004547093119395683,
      "loss": 4.7501,
      "mean_token_accuracy": 0.2181149899959564,
      "num_tokens": 46732840.0,
      "step": 25400
    },
    {
      "entropy": 5.041126537322998,
      "epoch": 2.122034747744738,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045469139721312607,
      "loss": 4.7307,
      "mean_token_accuracy": 0.2309171289205551,
      "num_tokens": 46742483.0,
      "step": 25405
    },
    {
      "entropy": 4.9593432426452635,
      "epoch": 2.1224523889074507,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004546734793409697,
      "loss": 4.6836,
      "mean_token_accuracy": 0.2256506562232971,
      "num_tokens": 46751217.0,
      "step": 25410
    },
    {
      "entropy": 5.078215169906616,
      "epoch": 2.1228700300701635,
      "grad_norm": 1.125,
      "learning_rate": 0.0004546555583234128,
      "loss": 4.7062,
      "mean_token_accuracy": 0.22532382756471633,
      "num_tokens": 46759978.0,
      "step": 25415
    },
    {
      "entropy": 5.045588111877441,
      "epoch": 2.1232876712328768,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000454637634160769,
      "loss": 4.748,
      "mean_token_accuracy": 0.2173048049211502,
      "num_tokens": 46768635.0,
      "step": 25420
    },
    {
      "entropy": 5.092543983459473,
      "epoch": 2.1237053123955896,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004546197068533523,
      "loss": 4.8598,
      "mean_token_accuracy": 0.20871696323156358,
      "num_tokens": 46778123.0,
      "step": 25425
    },
    {
      "entropy": 5.145368766784668,
      "epoch": 2.124122953558303,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004546017764014764,
      "loss": 4.8163,
      "mean_token_accuracy": 0.21840200275182725,
      "num_tokens": 46787821.0,
      "step": 25430
    },
    {
      "entropy": 4.987196731567383,
      "epoch": 2.1245405947210156,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004545838428054552,
      "loss": 4.7345,
      "mean_token_accuracy": 0.22444678843021393,
      "num_tokens": 46796419.0,
      "step": 25435
    },
    {
      "entropy": 5.105242395401001,
      "epoch": 2.124958235883729,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004545659060656027,
      "loss": 4.7147,
      "mean_token_accuracy": 0.21811379939317704,
      "num_tokens": 46805342.0,
      "step": 25440
    },
    {
      "entropy": 5.217288064956665,
      "epoch": 2.1253758770464417,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004545479661822328,
      "loss": 4.8287,
      "mean_token_accuracy": 0.21827236711978912,
      "num_tokens": 46813432.0,
      "step": 25445
    },
    {
      "entropy": 5.0383086681365965,
      "epoch": 2.1257935182091545,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00045453002315565965,
      "loss": 4.8579,
      "mean_token_accuracy": 0.2144441321492195,
      "num_tokens": 46823023.0,
      "step": 25450
    },
    {
      "entropy": 5.045850944519043,
      "epoch": 2.1262111593718678,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004545120769861974,
      "loss": 4.7541,
      "mean_token_accuracy": 0.22696213573217391,
      "num_tokens": 46832187.0,
      "step": 25455
    },
    {
      "entropy": 5.070642471313477,
      "epoch": 2.1266288005345806,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004544941276741601,
      "loss": 4.7623,
      "mean_token_accuracy": 0.22303605377674102,
      "num_tokens": 46840122.0,
      "step": 25460
    },
    {
      "entropy": 5.0725829124450685,
      "epoch": 2.127046441697294,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045447617521986205,
      "loss": 4.736,
      "mean_token_accuracy": 0.22284206449985505,
      "num_tokens": 46848854.0,
      "step": 25465
    },
    {
      "entropy": 5.119237041473388,
      "epoch": 2.1274640828600067,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004544582196236175,
      "loss": 4.8013,
      "mean_token_accuracy": 0.21631857454776765,
      "num_tokens": 46859182.0,
      "step": 25470
    },
    {
      "entropy": 4.986991930007934,
      "epoch": 2.1278817240227195,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004544402608857409,
      "loss": 4.7052,
      "mean_token_accuracy": 0.22389745861291885,
      "num_tokens": 46867856.0,
      "step": 25475
    },
    {
      "entropy": 4.999632930755615,
      "epoch": 2.1282993651854327,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004544222990065464,
      "loss": 4.7574,
      "mean_token_accuracy": 0.2199381783604622,
      "num_tokens": 46876793.0,
      "step": 25480
    },
    {
      "entropy": 4.981573534011841,
      "epoch": 2.1287170063481455,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045440433398634866,
      "loss": 4.7272,
      "mean_token_accuracy": 0.22636143565177919,
      "num_tokens": 46885342.0,
      "step": 25485
    },
    {
      "entropy": 5.098213863372803,
      "epoch": 2.129134647510859,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004543863658254622,
      "loss": 4.8056,
      "mean_token_accuracy": 0.21556583791971207,
      "num_tokens": 46895742.0,
      "step": 25490
    },
    {
      "entropy": 5.142532777786255,
      "epoch": 2.1295522886735716,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004543683945242013,
      "loss": 4.7529,
      "mean_token_accuracy": 0.22028662115335465,
      "num_tokens": 46905130.0,
      "step": 25495
    },
    {
      "entropy": 4.989952278137207,
      "epoch": 2.129969929836285,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045435042008288093,
      "loss": 4.7113,
      "mean_token_accuracy": 0.2206629753112793,
      "num_tokens": 46913939.0,
      "step": 25500
    },
    {
      "entropy": 5.058302164077759,
      "epoch": 2.1303875709989977,
      "grad_norm": 1.125,
      "learning_rate": 0.00045433244250181535,
      "loss": 4.7778,
      "mean_token_accuracy": 0.22209268510341645,
      "num_tokens": 46922308.0,
      "step": 25505
    },
    {
      "entropy": 5.0791534900665285,
      "epoch": 2.1308052121617105,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045431446178131966,
      "loss": 4.8682,
      "mean_token_accuracy": 0.22162727415561675,
      "num_tokens": 46931472.0,
      "step": 25510
    },
    {
      "entropy": 4.967384195327758,
      "epoch": 2.1312228533244237,
      "grad_norm": 1.125,
      "learning_rate": 0.0004542964779217085,
      "loss": 4.6828,
      "mean_token_accuracy": 0.2251712426543236,
      "num_tokens": 46940017.0,
      "step": 25515
    },
    {
      "entropy": 4.955354356765747,
      "epoch": 2.1316404944871366,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045427849092329655,
      "loss": 4.693,
      "mean_token_accuracy": 0.22171351164579392,
      "num_tokens": 46949045.0,
      "step": 25520
    },
    {
      "entropy": 5.068291044235229,
      "epoch": 2.13205813564985,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045426050078639884,
      "loss": 4.7352,
      "mean_token_accuracy": 0.2192837789654732,
      "num_tokens": 46957845.0,
      "step": 25525
    },
    {
      "entropy": 5.090725660324097,
      "epoch": 2.1324757768125626,
      "grad_norm": 1.125,
      "learning_rate": 0.0004542425075113303,
      "loss": 4.7042,
      "mean_token_accuracy": 0.22210424095392228,
      "num_tokens": 46966932.0,
      "step": 25530
    },
    {
      "entropy": 5.068354225158691,
      "epoch": 2.132893417975276,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045422451109840604,
      "loss": 4.7908,
      "mean_token_accuracy": 0.21541719436645507,
      "num_tokens": 46976362.0,
      "step": 25535
    },
    {
      "entropy": 5.030494165420532,
      "epoch": 2.1333110591379887,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045420651154794086,
      "loss": 4.7333,
      "mean_token_accuracy": 0.2215180516242981,
      "num_tokens": 46985311.0,
      "step": 25540
    },
    {
      "entropy": 5.141564607620239,
      "epoch": 2.1337287003007015,
      "grad_norm": 1.125,
      "learning_rate": 0.00045418850886025006,
      "loss": 4.7893,
      "mean_token_accuracy": 0.2141989901661873,
      "num_tokens": 46994666.0,
      "step": 25545
    },
    {
      "entropy": 5.026631498336792,
      "epoch": 2.1341463414634148,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045417050303564867,
      "loss": 4.7264,
      "mean_token_accuracy": 0.2215337872505188,
      "num_tokens": 47003894.0,
      "step": 25550
    },
    {
      "entropy": 4.989430379867554,
      "epoch": 2.1345639826261276,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045415249407445196,
      "loss": 4.6633,
      "mean_token_accuracy": 0.22568612396717072,
      "num_tokens": 47012556.0,
      "step": 25555
    },
    {
      "entropy": 5.04173059463501,
      "epoch": 2.134981623788841,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004541344819769751,
      "loss": 4.7388,
      "mean_token_accuracy": 0.21806716471910476,
      "num_tokens": 47022085.0,
      "step": 25560
    },
    {
      "entropy": 5.0109799861907955,
      "epoch": 2.1353992649515536,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045411646674353365,
      "loss": 4.7008,
      "mean_token_accuracy": 0.22737554907798768,
      "num_tokens": 47030759.0,
      "step": 25565
    },
    {
      "entropy": 5.071917915344239,
      "epoch": 2.1358169061142664,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045409844837444283,
      "loss": 4.7589,
      "mean_token_accuracy": 0.2184234619140625,
      "num_tokens": 47039877.0,
      "step": 25570
    },
    {
      "entropy": 4.986321306228637,
      "epoch": 2.1362345472769797,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045408042687001806,
      "loss": 4.7378,
      "mean_token_accuracy": 0.2183938890695572,
      "num_tokens": 47049500.0,
      "step": 25575
    },
    {
      "entropy": 5.1036903858184814,
      "epoch": 2.1366521884396925,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004540624022305749,
      "loss": 4.7928,
      "mean_token_accuracy": 0.2153424397110939,
      "num_tokens": 47059377.0,
      "step": 25580
    },
    {
      "entropy": 5.09178991317749,
      "epoch": 2.1370698296024058,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004540443744564288,
      "loss": 4.7727,
      "mean_token_accuracy": 0.22458950132131578,
      "num_tokens": 47068263.0,
      "step": 25585
    },
    {
      "entropy": 5.172622156143189,
      "epoch": 2.1374874707651186,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004540263435478953,
      "loss": 4.9055,
      "mean_token_accuracy": 0.20932086259126664,
      "num_tokens": 47077752.0,
      "step": 25590
    },
    {
      "entropy": 5.0838254451751705,
      "epoch": 2.1379051119278314,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045400830950529037,
      "loss": 4.7642,
      "mean_token_accuracy": 0.21949064880609512,
      "num_tokens": 47087013.0,
      "step": 25595
    },
    {
      "entropy": 4.971625661849975,
      "epoch": 2.1383227530905446,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004539902723289294,
      "loss": 4.6753,
      "mean_token_accuracy": 0.2279390200972557,
      "num_tokens": 47096502.0,
      "step": 25600
    },
    {
      "entropy": 5.08817138671875,
      "epoch": 2.1387403942532575,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045397223201912827,
      "loss": 4.8456,
      "mean_token_accuracy": 0.22105942517518998,
      "num_tokens": 47105347.0,
      "step": 25605
    },
    {
      "entropy": 5.004769515991211,
      "epoch": 2.1391580354159707,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004539541885762028,
      "loss": 4.6676,
      "mean_token_accuracy": 0.2272852912545204,
      "num_tokens": 47114137.0,
      "step": 25610
    },
    {
      "entropy": 5.131270170211792,
      "epoch": 2.1395756765786835,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004539361420004689,
      "loss": 4.8084,
      "mean_token_accuracy": 0.2178113952279091,
      "num_tokens": 47123366.0,
      "step": 25615
    },
    {
      "entropy": 5.14244384765625,
      "epoch": 2.139993317741397,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045391809229224236,
      "loss": 4.7481,
      "mean_token_accuracy": 0.22196035236120223,
      "num_tokens": 47132129.0,
      "step": 25620
    },
    {
      "entropy": 5.039553451538086,
      "epoch": 2.1404109589041096,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004539000394518393,
      "loss": 4.8624,
      "mean_token_accuracy": 0.221598881483078,
      "num_tokens": 47142860.0,
      "step": 25625
    },
    {
      "entropy": 5.078233242034912,
      "epoch": 2.1408286000668224,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004538819834795757,
      "loss": 4.805,
      "mean_token_accuracy": 0.2233329728245735,
      "num_tokens": 47151942.0,
      "step": 25630
    },
    {
      "entropy": 5.097681903839112,
      "epoch": 2.1412462412295357,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045386392437576765,
      "loss": 4.7851,
      "mean_token_accuracy": 0.21560911536216737,
      "num_tokens": 47160770.0,
      "step": 25635
    },
    {
      "entropy": 5.004749822616577,
      "epoch": 2.1416638823922485,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045384586214073134,
      "loss": 4.7179,
      "mean_token_accuracy": 0.22079337686300277,
      "num_tokens": 47169514.0,
      "step": 25640
    },
    {
      "entropy": 5.049567365646363,
      "epoch": 2.1420815235549617,
      "grad_norm": 1.125,
      "learning_rate": 0.00045382779677478294,
      "loss": 4.7395,
      "mean_token_accuracy": 0.22333415299654008,
      "num_tokens": 47178470.0,
      "step": 25645
    },
    {
      "entropy": 4.96357536315918,
      "epoch": 2.1424991647176745,
      "grad_norm": 1.125,
      "learning_rate": 0.00045380972827823866,
      "loss": 4.647,
      "mean_token_accuracy": 0.22942317128181458,
      "num_tokens": 47187694.0,
      "step": 25650
    },
    {
      "entropy": 4.996400785446167,
      "epoch": 2.142916805880388,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004537916566514149,
      "loss": 4.6864,
      "mean_token_accuracy": 0.22298550456762314,
      "num_tokens": 47196205.0,
      "step": 25655
    },
    {
      "entropy": 5.043460178375244,
      "epoch": 2.1433344470431006,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004537735818946279,
      "loss": 4.6873,
      "mean_token_accuracy": 0.22466426342725754,
      "num_tokens": 47205834.0,
      "step": 25660
    },
    {
      "entropy": 5.010080623626709,
      "epoch": 2.1437520882058134,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004537555040081942,
      "loss": 4.8474,
      "mean_token_accuracy": 0.21804654747247695,
      "num_tokens": 47215763.0,
      "step": 25665
    },
    {
      "entropy": 5.1155318260192875,
      "epoch": 2.1441697293685267,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045373742299243026,
      "loss": 4.8328,
      "mean_token_accuracy": 0.21300700604915618,
      "num_tokens": 47225536.0,
      "step": 25670
    },
    {
      "entropy": 5.110786294937133,
      "epoch": 2.1445873705312395,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045371933884765246,
      "loss": 4.7895,
      "mean_token_accuracy": 0.20951832383871077,
      "num_tokens": 47235104.0,
      "step": 25675
    },
    {
      "entropy": 5.0759971141815186,
      "epoch": 2.1450050116939527,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004537012515741777,
      "loss": 4.7538,
      "mean_token_accuracy": 0.21760960668325424,
      "num_tokens": 47244312.0,
      "step": 25680
    },
    {
      "entropy": 5.017222881317139,
      "epoch": 2.1454226528566656,
      "grad_norm": 1.125,
      "learning_rate": 0.0004536831611723223,
      "loss": 4.7776,
      "mean_token_accuracy": 0.21784751266241073,
      "num_tokens": 47253174.0,
      "step": 25685
    },
    {
      "entropy": 5.079057836532593,
      "epoch": 2.1458402940193784,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004536650676424032,
      "loss": 4.7409,
      "mean_token_accuracy": 0.22483112365007402,
      "num_tokens": 47262742.0,
      "step": 25690
    },
    {
      "entropy": 5.137144565582275,
      "epoch": 2.1462579351820916,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004536469709847369,
      "loss": 4.7975,
      "mean_token_accuracy": 0.21928559094667435,
      "num_tokens": 47271667.0,
      "step": 25695
    },
    {
      "entropy": 5.018533563613891,
      "epoch": 2.1466755763448044,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004536288711996404,
      "loss": 4.6699,
      "mean_token_accuracy": 0.22013687789440156,
      "num_tokens": 47280462.0,
      "step": 25700
    },
    {
      "entropy": 5.1061314105987545,
      "epoch": 2.1470932175075177,
      "grad_norm": 1.25,
      "learning_rate": 0.00045361076828743054,
      "loss": 4.8291,
      "mean_token_accuracy": 0.21472712606191635,
      "num_tokens": 47290834.0,
      "step": 25705
    },
    {
      "entropy": 5.056562232971191,
      "epoch": 2.1475108586702305,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004535926622484242,
      "loss": 4.7284,
      "mean_token_accuracy": 0.22652310281991958,
      "num_tokens": 47300002.0,
      "step": 25710
    },
    {
      "entropy": 5.144890880584716,
      "epoch": 2.1479284998329433,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045357455308293816,
      "loss": 4.8301,
      "mean_token_accuracy": 0.2226227790117264,
      "num_tokens": 47309299.0,
      "step": 25715
    },
    {
      "entropy": 5.023997163772583,
      "epoch": 2.1483461409956566,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004535564407912898,
      "loss": 4.7401,
      "mean_token_accuracy": 0.2191516250371933,
      "num_tokens": 47318680.0,
      "step": 25720
    },
    {
      "entropy": 5.045455455780029,
      "epoch": 2.1487637821583694,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004535383253737959,
      "loss": 4.7607,
      "mean_token_accuracy": 0.21530737727880478,
      "num_tokens": 47329841.0,
      "step": 25725
    },
    {
      "entropy": 5.127924871444702,
      "epoch": 2.1491814233210826,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004535202068307737,
      "loss": 4.8078,
      "mean_token_accuracy": 0.21538502126932144,
      "num_tokens": 47338962.0,
      "step": 25730
    },
    {
      "entropy": 5.088233709335327,
      "epoch": 2.1495990644837955,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004535020851625405,
      "loss": 4.7951,
      "mean_token_accuracy": 0.21454478204250335,
      "num_tokens": 47348230.0,
      "step": 25735
    },
    {
      "entropy": 5.066414022445679,
      "epoch": 2.1500167056465087,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045348396036941336,
      "loss": 4.8017,
      "mean_token_accuracy": 0.21138651221990584,
      "num_tokens": 47358694.0,
      "step": 25740
    },
    {
      "entropy": 5.084366512298584,
      "epoch": 2.1504343468092215,
      "grad_norm": 1.125,
      "learning_rate": 0.0004534658324517097,
      "loss": 4.7489,
      "mean_token_accuracy": 0.21561667770147325,
      "num_tokens": 47368617.0,
      "step": 25745
    },
    {
      "entropy": 5.14620270729065,
      "epoch": 2.1508519879719343,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004534477014097468,
      "loss": 4.8652,
      "mean_token_accuracy": 0.21302897334098816,
      "num_tokens": 47377902.0,
      "step": 25750
    },
    {
      "entropy": 5.082032346725464,
      "epoch": 2.1512696291346476,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045342956724384205,
      "loss": 4.7883,
      "mean_token_accuracy": 0.2286500573158264,
      "num_tokens": 47386337.0,
      "step": 25755
    },
    {
      "entropy": 5.046716737747192,
      "epoch": 2.1516872702973604,
      "grad_norm": 1.25,
      "learning_rate": 0.000453411429954313,
      "loss": 4.7441,
      "mean_token_accuracy": 0.22590435296297073,
      "num_tokens": 47395212.0,
      "step": 25760
    },
    {
      "entropy": 5.059244203567505,
      "epoch": 2.1521049114600737,
      "grad_norm": 1.25,
      "learning_rate": 0.0004533932895414771,
      "loss": 4.7652,
      "mean_token_accuracy": 0.22031215131282805,
      "num_tokens": 47404296.0,
      "step": 25765
    },
    {
      "entropy": 5.085462045669556,
      "epoch": 2.1525225526227865,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045337514600565196,
      "loss": 4.6812,
      "mean_token_accuracy": 0.2275671035051346,
      "num_tokens": 47413858.0,
      "step": 25770
    },
    {
      "entropy": 5.04687876701355,
      "epoch": 2.1529401937854997,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045335699934715513,
      "loss": 4.7561,
      "mean_token_accuracy": 0.21567681282758713,
      "num_tokens": 47422502.0,
      "step": 25775
    },
    {
      "entropy": 5.0403858661651615,
      "epoch": 2.1533578349482125,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045333884956630437,
      "loss": 4.7262,
      "mean_token_accuracy": 0.22431975454092026,
      "num_tokens": 47431763.0,
      "step": 25780
    },
    {
      "entropy": 5.174886560440063,
      "epoch": 2.1537754761109253,
      "grad_norm": 1.25,
      "learning_rate": 0.00045332069666341733,
      "loss": 4.9012,
      "mean_token_accuracy": 0.21678590178489685,
      "num_tokens": 47441261.0,
      "step": 25785
    },
    {
      "entropy": 5.066869163513184,
      "epoch": 2.1541931172736386,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004533025406388119,
      "loss": 4.8111,
      "mean_token_accuracy": 0.21685196906328202,
      "num_tokens": 47450028.0,
      "step": 25790
    },
    {
      "entropy": 5.1461607933044435,
      "epoch": 2.1546107584363514,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004532843814928059,
      "loss": 4.8333,
      "mean_token_accuracy": 0.220807945728302,
      "num_tokens": 47459617.0,
      "step": 25795
    },
    {
      "entropy": 5.080724382400513,
      "epoch": 2.1550283995990647,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004532662192257172,
      "loss": 4.6957,
      "mean_token_accuracy": 0.22533903270959854,
      "num_tokens": 47468748.0,
      "step": 25800
    },
    {
      "entropy": 5.053273868560791,
      "epoch": 2.1554460407617775,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004532480538378637,
      "loss": 4.6832,
      "mean_token_accuracy": 0.22268176823854446,
      "num_tokens": 47477294.0,
      "step": 25805
    },
    {
      "entropy": 5.071204137802124,
      "epoch": 2.1558636819244903,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045322988532956344,
      "loss": 4.8348,
      "mean_token_accuracy": 0.21222539395093917,
      "num_tokens": 47486733.0,
      "step": 25810
    },
    {
      "entropy": 5.090188932418823,
      "epoch": 2.1562813230872035,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045321171370113456,
      "loss": 4.738,
      "mean_token_accuracy": 0.21484940946102143,
      "num_tokens": 47495472.0,
      "step": 25815
    },
    {
      "entropy": 5.092970895767212,
      "epoch": 2.1566989642499164,
      "grad_norm": 1.25,
      "learning_rate": 0.0004531935389528951,
      "loss": 4.8235,
      "mean_token_accuracy": 0.21730822771787645,
      "num_tokens": 47505048.0,
      "step": 25820
    },
    {
      "entropy": 5.035326623916626,
      "epoch": 2.1571166054126296,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004531753610851633,
      "loss": 4.7398,
      "mean_token_accuracy": 0.22501375824213027,
      "num_tokens": 47513635.0,
      "step": 25825
    },
    {
      "entropy": 4.960976409912109,
      "epoch": 2.1575342465753424,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004531571800982573,
      "loss": 4.6768,
      "mean_token_accuracy": 0.22642754018306732,
      "num_tokens": 47522861.0,
      "step": 25830
    },
    {
      "entropy": 5.118049430847168,
      "epoch": 2.1579518877380552,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045313899599249536,
      "loss": 4.8469,
      "mean_token_accuracy": 0.21236506700515748,
      "num_tokens": 47531986.0,
      "step": 25835
    },
    {
      "entropy": 5.08786563873291,
      "epoch": 2.1583695289007685,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045312080876819597,
      "loss": 4.7904,
      "mean_token_accuracy": 0.20798466801643373,
      "num_tokens": 47542140.0,
      "step": 25840
    },
    {
      "entropy": 4.974657201766968,
      "epoch": 2.1587871700634813,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004531026184256773,
      "loss": 4.6356,
      "mean_token_accuracy": 0.2308322161436081,
      "num_tokens": 47550921.0,
      "step": 25845
    },
    {
      "entropy": 5.017920732498169,
      "epoch": 2.1592048112261946,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000453084424965258,
      "loss": 4.6199,
      "mean_token_accuracy": 0.23506717681884765,
      "num_tokens": 47560286.0,
      "step": 25850
    },
    {
      "entropy": 5.164158344268799,
      "epoch": 2.1596224523889074,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045306622838725644,
      "loss": 4.893,
      "mean_token_accuracy": 0.20600810647010803,
      "num_tokens": 47572089.0,
      "step": 25855
    },
    {
      "entropy": 5.075010347366333,
      "epoch": 2.1600400935516206,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004530480286919912,
      "loss": 4.7515,
      "mean_token_accuracy": 0.21755661070346832,
      "num_tokens": 47581602.0,
      "step": 25860
    },
    {
      "entropy": 5.0403382778167725,
      "epoch": 2.1604577347143334,
      "grad_norm": 1.0625,
      "learning_rate": 0.000453029825879781,
      "loss": 4.7422,
      "mean_token_accuracy": 0.22538864910602568,
      "num_tokens": 47591138.0,
      "step": 25865
    },
    {
      "entropy": 4.989988899230957,
      "epoch": 2.1608753758770463,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045301161995094433,
      "loss": 4.676,
      "mean_token_accuracy": 0.22846069782972336,
      "num_tokens": 47600182.0,
      "step": 25870
    },
    {
      "entropy": 5.157532548904419,
      "epoch": 2.1612930170397595,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045299341090579995,
      "loss": 4.9102,
      "mean_token_accuracy": 0.21204292476177217,
      "num_tokens": 47610207.0,
      "step": 25875
    },
    {
      "entropy": 5.076077556610107,
      "epoch": 2.1617106582024723,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004529751987446667,
      "loss": 4.7257,
      "mean_token_accuracy": 0.22229951918125151,
      "num_tokens": 47620276.0,
      "step": 25880
    },
    {
      "entropy": 5.017019987106323,
      "epoch": 2.1621282993651856,
      "grad_norm": 1.125,
      "learning_rate": 0.0004529569834678633,
      "loss": 4.8554,
      "mean_token_accuracy": 0.21536771357059478,
      "num_tokens": 47630187.0,
      "step": 25885
    },
    {
      "entropy": 5.097739315032959,
      "epoch": 2.1625459405278984,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004529387650757088,
      "loss": 4.7502,
      "mean_token_accuracy": 0.21865093111991882,
      "num_tokens": 47639187.0,
      "step": 25890
    },
    {
      "entropy": 5.041637802124024,
      "epoch": 2.1629635816906116,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045292054356852195,
      "loss": 4.7121,
      "mean_token_accuracy": 0.22303598374128342,
      "num_tokens": 47649272.0,
      "step": 25895
    },
    {
      "entropy": 4.9735006332397464,
      "epoch": 2.1633812228533245,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004529023189466219,
      "loss": 4.611,
      "mean_token_accuracy": 0.23816581070423126,
      "num_tokens": 47658046.0,
      "step": 25900
    },
    {
      "entropy": 4.901752471923828,
      "epoch": 2.1637988640160373,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045288409121032756,
      "loss": 4.6014,
      "mean_token_accuracy": 0.23829286247491838,
      "num_tokens": 47666658.0,
      "step": 25905
    },
    {
      "entropy": 5.055192232131958,
      "epoch": 2.1642165051787505,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045286586035995813,
      "loss": 4.7131,
      "mean_token_accuracy": 0.22081357836723328,
      "num_tokens": 47676455.0,
      "step": 25910
    },
    {
      "entropy": 5.031487989425659,
      "epoch": 2.1646341463414633,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045284762639583265,
      "loss": 4.6887,
      "mean_token_accuracy": 0.2221846967935562,
      "num_tokens": 47685518.0,
      "step": 25915
    },
    {
      "entropy": 4.975410985946655,
      "epoch": 2.1650517875041766,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045282938931827046,
      "loss": 4.7649,
      "mean_token_accuracy": 0.22396631091833114,
      "num_tokens": 47695070.0,
      "step": 25920
    },
    {
      "entropy": 5.081592655181884,
      "epoch": 2.1654694286668894,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004528111491275907,
      "loss": 4.7693,
      "mean_token_accuracy": 0.22078019082546235,
      "num_tokens": 47704650.0,
      "step": 25925
    },
    {
      "entropy": 5.082094144821167,
      "epoch": 2.165887069829602,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004527929058241128,
      "loss": 4.7918,
      "mean_token_accuracy": 0.2230933278799057,
      "num_tokens": 47714263.0,
      "step": 25930
    },
    {
      "entropy": 5.087661790847778,
      "epoch": 2.1663047109923155,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045277465940815603,
      "loss": 4.8039,
      "mean_token_accuracy": 0.22031426727771758,
      "num_tokens": 47723448.0,
      "step": 25935
    },
    {
      "entropy": 5.008620548248291,
      "epoch": 2.1667223521550283,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004527564098800399,
      "loss": 4.7023,
      "mean_token_accuracy": 0.230794058740139,
      "num_tokens": 47731841.0,
      "step": 25940
    },
    {
      "entropy": 4.975513505935669,
      "epoch": 2.1671399933177415,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004527381572400838,
      "loss": 4.6222,
      "mean_token_accuracy": 0.22604460567235946,
      "num_tokens": 47741422.0,
      "step": 25945
    },
    {
      "entropy": 4.975473785400391,
      "epoch": 2.1675576344804544,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045271990148860737,
      "loss": 4.6934,
      "mean_token_accuracy": 0.21996889710426332,
      "num_tokens": 47750601.0,
      "step": 25950
    },
    {
      "entropy": 5.089600419998169,
      "epoch": 2.167975275643167,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045270164262593013,
      "loss": 4.8569,
      "mean_token_accuracy": 0.209135439991951,
      "num_tokens": 47759992.0,
      "step": 25955
    },
    {
      "entropy": 4.959321594238281,
      "epoch": 2.1683929168058804,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004526833806523717,
      "loss": 4.6458,
      "mean_token_accuracy": 0.21882169544696808,
      "num_tokens": 47769548.0,
      "step": 25960
    },
    {
      "entropy": 5.015884780883789,
      "epoch": 2.1688105579685932,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045266511556825184,
      "loss": 4.7797,
      "mean_token_accuracy": 0.2176513060927391,
      "num_tokens": 47778998.0,
      "step": 25965
    },
    {
      "entropy": 5.097615814208984,
      "epoch": 2.1692281991313065,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045264684737389026,
      "loss": 4.7543,
      "mean_token_accuracy": 0.21827871054410936,
      "num_tokens": 47787931.0,
      "step": 25970
    },
    {
      "entropy": 5.085480499267578,
      "epoch": 2.1696458402940193,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045262857606960684,
      "loss": 4.7398,
      "mean_token_accuracy": 0.22123963683843612,
      "num_tokens": 47797421.0,
      "step": 25975
    },
    {
      "entropy": 5.060890960693359,
      "epoch": 2.1700634814567326,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045261030165572134,
      "loss": 4.7984,
      "mean_token_accuracy": 0.2111775815486908,
      "num_tokens": 47807153.0,
      "step": 25980
    },
    {
      "entropy": 5.100312519073486,
      "epoch": 2.1704811226194454,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004525920241325537,
      "loss": 4.8232,
      "mean_token_accuracy": 0.21842971444129944,
      "num_tokens": 47816840.0,
      "step": 25985
    },
    {
      "entropy": 5.147786808013916,
      "epoch": 2.170898763782158,
      "grad_norm": 1.125,
      "learning_rate": 0.0004525737435004239,
      "loss": 4.8215,
      "mean_token_accuracy": 0.21974510550498963,
      "num_tokens": 47825997.0,
      "step": 25990
    },
    {
      "entropy": 5.071097421646118,
      "epoch": 2.1713164049448714,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000452555459759652,
      "loss": 4.7412,
      "mean_token_accuracy": 0.2222074404358864,
      "num_tokens": 47834106.0,
      "step": 25995
    },
    {
      "entropy": 5.0085550308227536,
      "epoch": 2.1717340461075842,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004525371729105581,
      "loss": 4.7016,
      "mean_token_accuracy": 0.23367587327957154,
      "num_tokens": 47843161.0,
      "step": 26000
    },
    {
      "entropy": 5.016199684143066,
      "epoch": 2.1721516872702975,
      "grad_norm": 1.125,
      "learning_rate": 0.0004525188829534622,
      "loss": 4.6825,
      "mean_token_accuracy": 0.22441954016685486,
      "num_tokens": 47852396.0,
      "step": 26005
    },
    {
      "entropy": 5.022689294815064,
      "epoch": 2.1725693284330103,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004525005898886847,
      "loss": 4.7738,
      "mean_token_accuracy": 0.22584887742996215,
      "num_tokens": 47861070.0,
      "step": 26010
    },
    {
      "entropy": 5.045599174499512,
      "epoch": 2.1729869695957236,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004524822937165456,
      "loss": 4.8022,
      "mean_token_accuracy": 0.21298011690378188,
      "num_tokens": 47869771.0,
      "step": 26015
    },
    {
      "entropy": 5.111841773986816,
      "epoch": 2.1734046107584364,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045246399443736536,
      "loss": 4.8333,
      "mean_token_accuracy": 0.21939531862735748,
      "num_tokens": 47879447.0,
      "step": 26020
    },
    {
      "entropy": 5.085912561416626,
      "epoch": 2.173822251921149,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045244569205146425,
      "loss": 4.7323,
      "mean_token_accuracy": 0.22399126440286637,
      "num_tokens": 47887415.0,
      "step": 26025
    },
    {
      "entropy": 5.120169925689697,
      "epoch": 2.1742398930838625,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004524273865591627,
      "loss": 4.8558,
      "mean_token_accuracy": 0.21420131027698516,
      "num_tokens": 47896898.0,
      "step": 26030
    },
    {
      "entropy": 5.037479066848755,
      "epoch": 2.1746575342465753,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045240907796078125,
      "loss": 4.7662,
      "mean_token_accuracy": 0.22477283775806428,
      "num_tokens": 47906453.0,
      "step": 26035
    },
    {
      "entropy": 5.114332675933838,
      "epoch": 2.1750751754092885,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004523907662566403,
      "loss": 4.8218,
      "mean_token_accuracy": 0.21470576226711274,
      "num_tokens": 47916614.0,
      "step": 26040
    },
    {
      "entropy": 5.078053903579712,
      "epoch": 2.1754928165720013,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045237245144706044,
      "loss": 4.766,
      "mean_token_accuracy": 0.21895987987518312,
      "num_tokens": 47925924.0,
      "step": 26045
    },
    {
      "entropy": 5.116511678695678,
      "epoch": 2.175910457734714,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004523541335323624,
      "loss": 4.8178,
      "mean_token_accuracy": 0.21438489109277725,
      "num_tokens": 47934706.0,
      "step": 26050
    },
    {
      "entropy": 5.1206964492797855,
      "epoch": 2.1763280988974274,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004523358125128667,
      "loss": 4.8674,
      "mean_token_accuracy": 0.21052806675434113,
      "num_tokens": 47944885.0,
      "step": 26055
    },
    {
      "entropy": 5.024005365371704,
      "epoch": 2.17674574006014,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004523174883888941,
      "loss": 4.735,
      "mean_token_accuracy": 0.2242010712623596,
      "num_tokens": 47953491.0,
      "step": 26060
    },
    {
      "entropy": 5.0773231983184814,
      "epoch": 2.1771633812228535,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004522991611607655,
      "loss": 4.8339,
      "mean_token_accuracy": 0.21183476001024246,
      "num_tokens": 47962583.0,
      "step": 26065
    },
    {
      "entropy": 5.092675018310547,
      "epoch": 2.1775810223855663,
      "grad_norm": 1.125,
      "learning_rate": 0.00045228083082880164,
      "loss": 4.7151,
      "mean_token_accuracy": 0.22425554245710372,
      "num_tokens": 47971213.0,
      "step": 26070
    },
    {
      "entropy": 5.159971475601196,
      "epoch": 2.177998663548279,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004522624973933234,
      "loss": 4.9255,
      "mean_token_accuracy": 0.20696061104536057,
      "num_tokens": 47981001.0,
      "step": 26075
    },
    {
      "entropy": 5.062496709823608,
      "epoch": 2.1784163047109923,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045224416085465195,
      "loss": 4.7653,
      "mean_token_accuracy": 0.21928724348545076,
      "num_tokens": 47989765.0,
      "step": 26080
    },
    {
      "entropy": 5.122473907470703,
      "epoch": 2.178833945873705,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045222582121310793,
      "loss": 4.7491,
      "mean_token_accuracy": 0.22582947462797165,
      "num_tokens": 47999676.0,
      "step": 26085
    },
    {
      "entropy": 5.063970565795898,
      "epoch": 2.1792515870364184,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045220747846901266,
      "loss": 4.7654,
      "mean_token_accuracy": 0.216520956158638,
      "num_tokens": 48009181.0,
      "step": 26090
    },
    {
      "entropy": 4.920602750778198,
      "epoch": 2.1796692281991312,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045218913262268717,
      "loss": 4.6183,
      "mean_token_accuracy": 0.23771039694547652,
      "num_tokens": 48018436.0,
      "step": 26095
    },
    {
      "entropy": 5.056920146942138,
      "epoch": 2.1800868693618445,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045217078367445264,
      "loss": 4.7953,
      "mean_token_accuracy": 0.2198444664478302,
      "num_tokens": 48027527.0,
      "step": 26100
    },
    {
      "entropy": 5.099845552444458,
      "epoch": 2.1805045105245573,
      "grad_norm": 1.015625,
      "learning_rate": 0.00045215243162463023,
      "loss": 4.7317,
      "mean_token_accuracy": 0.2255525454878807,
      "num_tokens": 48037432.0,
      "step": 26105
    },
    {
      "entropy": 5.051981019973755,
      "epoch": 2.18092215168727,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004521340764735413,
      "loss": 4.8147,
      "mean_token_accuracy": 0.21644293516874313,
      "num_tokens": 48046037.0,
      "step": 26110
    },
    {
      "entropy": 5.0303867816925045,
      "epoch": 2.1813397928499834,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004521157182215071,
      "loss": 4.7252,
      "mean_token_accuracy": 0.21527319848537446,
      "num_tokens": 48055197.0,
      "step": 26115
    },
    {
      "entropy": 5.154839134216308,
      "epoch": 2.181757434012696,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045209735686884913,
      "loss": 4.9482,
      "mean_token_accuracy": 0.19972267299890517,
      "num_tokens": 48064152.0,
      "step": 26120
    },
    {
      "entropy": 5.067800903320313,
      "epoch": 2.1821750751754094,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045207899241588874,
      "loss": 4.7424,
      "mean_token_accuracy": 0.2256841018795967,
      "num_tokens": 48073765.0,
      "step": 26125
    },
    {
      "entropy": 5.053808307647705,
      "epoch": 2.1825927163381222,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004520606248629474,
      "loss": 4.6818,
      "mean_token_accuracy": 0.23072569966316223,
      "num_tokens": 48082816.0,
      "step": 26130
    },
    {
      "entropy": 5.0996153354644775,
      "epoch": 2.1830103575008355,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004520422542103467,
      "loss": 4.7666,
      "mean_token_accuracy": 0.22160764634609223,
      "num_tokens": 48091841.0,
      "step": 26135
    },
    {
      "entropy": 4.961469650268555,
      "epoch": 2.1834279986635483,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004520238804584083,
      "loss": 4.6818,
      "mean_token_accuracy": 0.22169484347105026,
      "num_tokens": 48102162.0,
      "step": 26140
    },
    {
      "entropy": 5.068823146820068,
      "epoch": 2.183845639826261,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004520055036074538,
      "loss": 4.738,
      "mean_token_accuracy": 0.22687319070100784,
      "num_tokens": 48111092.0,
      "step": 26145
    },
    {
      "entropy": 4.960451459884643,
      "epoch": 2.1842632809889744,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004519871236578048,
      "loss": 4.6385,
      "mean_token_accuracy": 0.22330132722854615,
      "num_tokens": 48119643.0,
      "step": 26150
    },
    {
      "entropy": 4.982569026947021,
      "epoch": 2.184680922151687,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004519687406097832,
      "loss": 4.6791,
      "mean_token_accuracy": 0.2258937180042267,
      "num_tokens": 48128685.0,
      "step": 26155
    },
    {
      "entropy": 5.005975294113159,
      "epoch": 2.1850985633144004,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004519503544637107,
      "loss": 4.7159,
      "mean_token_accuracy": 0.2277661845088005,
      "num_tokens": 48137645.0,
      "step": 26160
    },
    {
      "entropy": 5.0853088855743405,
      "epoch": 2.1855162044771133,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045193196521990935,
      "loss": 4.7883,
      "mean_token_accuracy": 0.21970328986644744,
      "num_tokens": 48147240.0,
      "step": 26165
    },
    {
      "entropy": 5.044320106506348,
      "epoch": 2.185933845639826,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000451913572878701,
      "loss": 4.7917,
      "mean_token_accuracy": 0.21503681391477586,
      "num_tokens": 48156992.0,
      "step": 26170
    },
    {
      "entropy": 5.169263458251953,
      "epoch": 2.1863514868025393,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004518951774404076,
      "loss": 4.804,
      "mean_token_accuracy": 0.2155632734298706,
      "num_tokens": 48165864.0,
      "step": 26175
    },
    {
      "entropy": 5.119516229629516,
      "epoch": 2.186769127965252,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004518767789053511,
      "loss": 4.8189,
      "mean_token_accuracy": 0.22061990052461625,
      "num_tokens": 48174676.0,
      "step": 26180
    },
    {
      "entropy": 4.957622098922729,
      "epoch": 2.1871867691279654,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045185837727385373,
      "loss": 4.6189,
      "mean_token_accuracy": 0.2312345862388611,
      "num_tokens": 48182838.0,
      "step": 26185
    },
    {
      "entropy": 5.120576333999634,
      "epoch": 2.187604410290678,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004518399725462376,
      "loss": 4.8319,
      "mean_token_accuracy": 0.21238461881875992,
      "num_tokens": 48192087.0,
      "step": 26190
    },
    {
      "entropy": 4.992107295989991,
      "epoch": 2.188022051453391,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004518215647228248,
      "loss": 4.7076,
      "mean_token_accuracy": 0.22218371331691741,
      "num_tokens": 48201020.0,
      "step": 26195
    },
    {
      "entropy": 5.0643157958984375,
      "epoch": 2.1884396926161043,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045180315380393776,
      "loss": 4.7224,
      "mean_token_accuracy": 0.21985504031181335,
      "num_tokens": 48210672.0,
      "step": 26200
    },
    {
      "entropy": 5.083449840545654,
      "epoch": 2.188857333778817,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004517847397898987,
      "loss": 4.8013,
      "mean_token_accuracy": 0.2172952637076378,
      "num_tokens": 48220669.0,
      "step": 26205
    },
    {
      "entropy": 5.036480569839478,
      "epoch": 2.1892749749415303,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045176632268102994,
      "loss": 4.5913,
      "mean_token_accuracy": 0.24072306901216506,
      "num_tokens": 48230198.0,
      "step": 26210
    },
    {
      "entropy": 5.0601602554321286,
      "epoch": 2.189692616104243,
      "grad_norm": 1.140625,
      "learning_rate": 0.000451747902477654,
      "loss": 4.843,
      "mean_token_accuracy": 0.20493164360523225,
      "num_tokens": 48239854.0,
      "step": 26215
    },
    {
      "entropy": 5.054612302780152,
      "epoch": 2.1901102572669564,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045172947918009317,
      "loss": 4.8563,
      "mean_token_accuracy": 0.21320455372333527,
      "num_tokens": 48249239.0,
      "step": 26220
    },
    {
      "entropy": 5.089053440093994,
      "epoch": 2.190527898429669,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004517110527886701,
      "loss": 4.6924,
      "mean_token_accuracy": 0.22463549673557281,
      "num_tokens": 48258196.0,
      "step": 26225
    },
    {
      "entropy": 5.019190835952759,
      "epoch": 2.190945539592382,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004516926233037074,
      "loss": 4.6962,
      "mean_token_accuracy": 0.2205629825592041,
      "num_tokens": 48267102.0,
      "step": 26230
    },
    {
      "entropy": 5.072163200378418,
      "epoch": 2.1913631807550953,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004516741907255276,
      "loss": 4.769,
      "mean_token_accuracy": 0.2144907906651497,
      "num_tokens": 48276273.0,
      "step": 26235
    },
    {
      "entropy": 5.050119924545288,
      "epoch": 2.191780821917808,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004516557550544534,
      "loss": 4.7432,
      "mean_token_accuracy": 0.22194526046514512,
      "num_tokens": 48284921.0,
      "step": 26240
    },
    {
      "entropy": 5.005906486511231,
      "epoch": 2.1921984630805214,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004516373162908077,
      "loss": 4.784,
      "mean_token_accuracy": 0.2235051542520523,
      "num_tokens": 48293188.0,
      "step": 26245
    },
    {
      "entropy": 5.091838550567627,
      "epoch": 2.192616104243234,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045161887443491297,
      "loss": 4.7784,
      "mean_token_accuracy": 0.2191379889845848,
      "num_tokens": 48301898.0,
      "step": 26250
    },
    {
      "entropy": 5.070802497863769,
      "epoch": 2.1930337454059474,
      "grad_norm": 1.265625,
      "learning_rate": 0.00045160042948709235,
      "loss": 4.6738,
      "mean_token_accuracy": 0.23254324048757552,
      "num_tokens": 48310291.0,
      "step": 26255
    },
    {
      "entropy": 5.045658969879151,
      "epoch": 2.1934513865686602,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045158198144766863,
      "loss": 4.6715,
      "mean_token_accuracy": 0.23021074384450912,
      "num_tokens": 48318922.0,
      "step": 26260
    },
    {
      "entropy": 4.98188157081604,
      "epoch": 2.193869027731373,
      "grad_norm": 1.125,
      "learning_rate": 0.0004515635303169648,
      "loss": 4.7256,
      "mean_token_accuracy": 0.23093071281909944,
      "num_tokens": 48327843.0,
      "step": 26265
    },
    {
      "entropy": 5.0919441223144535,
      "epoch": 2.1942866688940863,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045154507609530375,
      "loss": 4.8113,
      "mean_token_accuracy": 0.22291814684867858,
      "num_tokens": 48336756.0,
      "step": 26270
    },
    {
      "entropy": 5.182994365692139,
      "epoch": 2.194704310056799,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004515266187830087,
      "loss": 4.8936,
      "mean_token_accuracy": 0.21054655760526658,
      "num_tokens": 48346813.0,
      "step": 26275
    },
    {
      "entropy": 5.089128494262695,
      "epoch": 2.1951219512195124,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045150815838040275,
      "loss": 4.8137,
      "mean_token_accuracy": 0.21570929288864135,
      "num_tokens": 48355416.0,
      "step": 26280
    },
    {
      "entropy": 5.210161781311035,
      "epoch": 2.195539592382225,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045148969488780893,
      "loss": 4.8854,
      "mean_token_accuracy": 0.21414125710725784,
      "num_tokens": 48364824.0,
      "step": 26285
    },
    {
      "entropy": 5.088964986801147,
      "epoch": 2.195957233544938,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004514712283055506,
      "loss": 4.7231,
      "mean_token_accuracy": 0.21665676087141036,
      "num_tokens": 48375016.0,
      "step": 26290
    },
    {
      "entropy": 4.981373834609985,
      "epoch": 2.1963748747076512,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004514527586339511,
      "loss": 4.6723,
      "mean_token_accuracy": 0.22809708416461943,
      "num_tokens": 48383232.0,
      "step": 26295
    },
    {
      "entropy": 5.067272329330445,
      "epoch": 2.196792515870364,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045143428587333354,
      "loss": 4.82,
      "mean_token_accuracy": 0.2150905102491379,
      "num_tokens": 48392374.0,
      "step": 26300
    },
    {
      "entropy": 5.127012729644775,
      "epoch": 2.1972101570330773,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004514158100240214,
      "loss": 4.8533,
      "mean_token_accuracy": 0.20964490324258805,
      "num_tokens": 48401471.0,
      "step": 26305
    },
    {
      "entropy": 5.18292269706726,
      "epoch": 2.19762779819579,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004513973310863383,
      "loss": 4.8417,
      "mean_token_accuracy": 0.21300113797187806,
      "num_tokens": 48410585.0,
      "step": 26310
    },
    {
      "entropy": 5.028117275238037,
      "epoch": 2.198045439358503,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045137884906060747,
      "loss": 4.7277,
      "mean_token_accuracy": 0.22132205069065095,
      "num_tokens": 48419563.0,
      "step": 26315
    },
    {
      "entropy": 5.004578447341919,
      "epoch": 2.198463080521216,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004513603639471526,
      "loss": 4.7112,
      "mean_token_accuracy": 0.22427908331155777,
      "num_tokens": 48429801.0,
      "step": 26320
    },
    {
      "entropy": 5.0767230033874515,
      "epoch": 2.198880721683929,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004513418757462973,
      "loss": 4.7117,
      "mean_token_accuracy": 0.2233252450823784,
      "num_tokens": 48437953.0,
      "step": 26325
    },
    {
      "entropy": 5.074353981018066,
      "epoch": 2.1992983628466423,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045132338445836516,
      "loss": 4.768,
      "mean_token_accuracy": 0.21748291999101638,
      "num_tokens": 48447733.0,
      "step": 26330
    },
    {
      "entropy": 5.025275754928589,
      "epoch": 2.199716004009355,
      "grad_norm": 1.125,
      "learning_rate": 0.00045130489008367997,
      "loss": 4.7166,
      "mean_token_accuracy": 0.22388099730014802,
      "num_tokens": 48456530.0,
      "step": 26335
    },
    {
      "entropy": 5.015954160690308,
      "epoch": 2.2001336451720683,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004512863926225655,
      "loss": 4.7291,
      "mean_token_accuracy": 0.22747136801481246,
      "num_tokens": 48465607.0,
      "step": 26340
    },
    {
      "entropy": 5.0403200626373295,
      "epoch": 2.200551286334781,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045126789207534544,
      "loss": 4.6769,
      "mean_token_accuracy": 0.22227276712656022,
      "num_tokens": 48474112.0,
      "step": 26345
    },
    {
      "entropy": 5.045609712600708,
      "epoch": 2.200968927497494,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045124938844234386,
      "loss": 4.7077,
      "mean_token_accuracy": 0.2315431907773018,
      "num_tokens": 48482677.0,
      "step": 26350
    },
    {
      "entropy": 5.000299787521362,
      "epoch": 2.201386568660207,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045123088172388455,
      "loss": 4.6742,
      "mean_token_accuracy": 0.23170193582773208,
      "num_tokens": 48491313.0,
      "step": 26355
    },
    {
      "entropy": 5.011080408096314,
      "epoch": 2.20180420982292,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045121237192029154,
      "loss": 4.7855,
      "mean_token_accuracy": 0.226299686729908,
      "num_tokens": 48501106.0,
      "step": 26360
    },
    {
      "entropy": 5.130573987960815,
      "epoch": 2.2022218509856333,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045119385903188877,
      "loss": 4.7754,
      "mean_token_accuracy": 0.22056301087141036,
      "num_tokens": 48511219.0,
      "step": 26365
    },
    {
      "entropy": 5.028763389587402,
      "epoch": 2.202639492148346,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004511753430590004,
      "loss": 4.7638,
      "mean_token_accuracy": 0.22911556214094161,
      "num_tokens": 48520436.0,
      "step": 26370
    },
    {
      "entropy": 5.040529489517212,
      "epoch": 2.2030571333110593,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004511568240019507,
      "loss": 4.7947,
      "mean_token_accuracy": 0.21619257479906082,
      "num_tokens": 48531516.0,
      "step": 26375
    },
    {
      "entropy": 5.080737352371216,
      "epoch": 2.203474774473772,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045113830186106367,
      "loss": 4.6693,
      "mean_token_accuracy": 0.22372309416532515,
      "num_tokens": 48540512.0,
      "step": 26380
    },
    {
      "entropy": 5.091280794143676,
      "epoch": 2.203892415636485,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004511197766366637,
      "loss": 4.8218,
      "mean_token_accuracy": 0.2107875257730484,
      "num_tokens": 48550273.0,
      "step": 26385
    },
    {
      "entropy": 5.001807069778442,
      "epoch": 2.2043100567991982,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045110124832907505,
      "loss": 4.7798,
      "mean_token_accuracy": 0.22681020498275756,
      "num_tokens": 48558606.0,
      "step": 26390
    },
    {
      "entropy": 5.0349452018737795,
      "epoch": 2.204727697961911,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004510827169386221,
      "loss": 4.7307,
      "mean_token_accuracy": 0.21686634719371795,
      "num_tokens": 48567167.0,
      "step": 26395
    },
    {
      "entropy": 5.018091869354248,
      "epoch": 2.2051453391246243,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004510641824656292,
      "loss": 4.7213,
      "mean_token_accuracy": 0.21722577810287474,
      "num_tokens": 48576275.0,
      "step": 26400
    },
    {
      "entropy": 5.048200464248657,
      "epoch": 2.205562980287337,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004510456449104208,
      "loss": 4.7135,
      "mean_token_accuracy": 0.22502057254314423,
      "num_tokens": 48585217.0,
      "step": 26405
    },
    {
      "entropy": 5.088835716247559,
      "epoch": 2.20598062145005,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004510271042733215,
      "loss": 4.736,
      "mean_token_accuracy": 0.2197560504078865,
      "num_tokens": 48594583.0,
      "step": 26410
    },
    {
      "entropy": 4.92223162651062,
      "epoch": 2.206398262612763,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004510085605546559,
      "loss": 4.6476,
      "mean_token_accuracy": 0.2346424490213394,
      "num_tokens": 48603677.0,
      "step": 26415
    },
    {
      "entropy": 5.107379817962647,
      "epoch": 2.206815903775476,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045099001375474855,
      "loss": 4.8207,
      "mean_token_accuracy": 0.2173387736082077,
      "num_tokens": 48614453.0,
      "step": 26420
    },
    {
      "entropy": 5.017702198028564,
      "epoch": 2.2072335449381892,
      "grad_norm": 1.125,
      "learning_rate": 0.00045097146387392423,
      "loss": 4.7427,
      "mean_token_accuracy": 0.22690527737140656,
      "num_tokens": 48622871.0,
      "step": 26425
    },
    {
      "entropy": 5.033454704284668,
      "epoch": 2.207651186100902,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045095291091250755,
      "loss": 4.6708,
      "mean_token_accuracy": 0.2221214458346367,
      "num_tokens": 48631466.0,
      "step": 26430
    },
    {
      "entropy": 5.006383466720581,
      "epoch": 2.2080688272636153,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004509343548708235,
      "loss": 4.6623,
      "mean_token_accuracy": 0.23399909287691117,
      "num_tokens": 48640020.0,
      "step": 26435
    },
    {
      "entropy": 4.996470022201538,
      "epoch": 2.208486468426328,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004509157957491966,
      "loss": 4.8034,
      "mean_token_accuracy": 0.22260961830615997,
      "num_tokens": 48648428.0,
      "step": 26440
    },
    {
      "entropy": 5.015123653411865,
      "epoch": 2.208904109589041,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045089723354795204,
      "loss": 4.6304,
      "mean_token_accuracy": 0.2299093261361122,
      "num_tokens": 48657232.0,
      "step": 26445
    },
    {
      "entropy": 4.959100246429443,
      "epoch": 2.209321750751754,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004508786682674147,
      "loss": 4.6303,
      "mean_token_accuracy": 0.2325364977121353,
      "num_tokens": 48665802.0,
      "step": 26450
    },
    {
      "entropy": 4.968886184692383,
      "epoch": 2.209739391914467,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045086009990790956,
      "loss": 4.6551,
      "mean_token_accuracy": 0.23181230574846268,
      "num_tokens": 48674622.0,
      "step": 26455
    },
    {
      "entropy": 4.991921472549438,
      "epoch": 2.2101570330771803,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045084152846976166,
      "loss": 4.7242,
      "mean_token_accuracy": 0.22630414813756944,
      "num_tokens": 48683365.0,
      "step": 26460
    },
    {
      "entropy": 5.070567846298218,
      "epoch": 2.210574674239893,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045082295395329616,
      "loss": 4.7212,
      "mean_token_accuracy": 0.22185125201940536,
      "num_tokens": 48693554.0,
      "step": 26465
    },
    {
      "entropy": 5.066930532455444,
      "epoch": 2.210992315402606,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004508043763588382,
      "loss": 4.7181,
      "mean_token_accuracy": 0.22514306157827377,
      "num_tokens": 48702892.0,
      "step": 26470
    },
    {
      "entropy": 5.035771799087525,
      "epoch": 2.211409956565319,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045078579568671307,
      "loss": 4.7638,
      "mean_token_accuracy": 0.21893573105335234,
      "num_tokens": 48712036.0,
      "step": 26475
    },
    {
      "entropy": 5.058572387695312,
      "epoch": 2.211827597728032,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004507672119372459,
      "loss": 4.7908,
      "mean_token_accuracy": 0.2194559320807457,
      "num_tokens": 48721462.0,
      "step": 26480
    },
    {
      "entropy": 5.048501205444336,
      "epoch": 2.212245238890745,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045074862511076215,
      "loss": 4.7454,
      "mean_token_accuracy": 0.21880227476358413,
      "num_tokens": 48730542.0,
      "step": 26485
    },
    {
      "entropy": 5.0257728099823,
      "epoch": 2.212662880053458,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004507300352075872,
      "loss": 4.7704,
      "mean_token_accuracy": 0.22346299290657043,
      "num_tokens": 48739477.0,
      "step": 26490
    },
    {
      "entropy": 5.1173804759979244,
      "epoch": 2.2130805212161713,
      "grad_norm": 1.125,
      "learning_rate": 0.00045071144222804646,
      "loss": 4.7726,
      "mean_token_accuracy": 0.22411481589078902,
      "num_tokens": 48748945.0,
      "step": 26495
    },
    {
      "entropy": 5.112411689758301,
      "epoch": 2.213498162378884,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004506928461724654,
      "loss": 4.7809,
      "mean_token_accuracy": 0.2212671607732773,
      "num_tokens": 48757762.0,
      "step": 26500
    },
    {
      "entropy": 5.199615621566773,
      "epoch": 2.213915803541597,
      "grad_norm": 1.125,
      "learning_rate": 0.00045067424704116956,
      "loss": 4.9401,
      "mean_token_accuracy": 0.21026334762573243,
      "num_tokens": 48767773.0,
      "step": 26505
    },
    {
      "entropy": 5.08559136390686,
      "epoch": 2.21433344470431,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004506556448344846,
      "loss": 4.8056,
      "mean_token_accuracy": 0.21962883323431015,
      "num_tokens": 48777499.0,
      "step": 26510
    },
    {
      "entropy": 5.051616477966308,
      "epoch": 2.214751085867023,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004506370395527362,
      "loss": 4.683,
      "mean_token_accuracy": 0.22426982522010802,
      "num_tokens": 48786499.0,
      "step": 26515
    },
    {
      "entropy": 5.023867177963257,
      "epoch": 2.215168727029736,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045061843119624984,
      "loss": 4.7827,
      "mean_token_accuracy": 0.2226831391453743,
      "num_tokens": 48795086.0,
      "step": 26520
    },
    {
      "entropy": 5.033061981201172,
      "epoch": 2.215586368192449,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004505998197653516,
      "loss": 4.6729,
      "mean_token_accuracy": 0.22184403389692306,
      "num_tokens": 48804131.0,
      "step": 26525
    },
    {
      "entropy": 5.098412370681762,
      "epoch": 2.216004009355162,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004505812052603671,
      "loss": 4.7816,
      "mean_token_accuracy": 0.21658101230859755,
      "num_tokens": 48812986.0,
      "step": 26530
    },
    {
      "entropy": 5.235426807403565,
      "epoch": 2.216421650517875,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004505625876816222,
      "loss": 4.8697,
      "mean_token_accuracy": 0.21709206104278564,
      "num_tokens": 48822200.0,
      "step": 26535
    },
    {
      "entropy": 5.178609180450439,
      "epoch": 2.216839291680588,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045054396702944293,
      "loss": 4.8005,
      "mean_token_accuracy": 0.22563101053237916,
      "num_tokens": 48831485.0,
      "step": 26540
    },
    {
      "entropy": 4.891719579696655,
      "epoch": 2.217256932843301,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045052534330415525,
      "loss": 4.5843,
      "mean_token_accuracy": 0.2330525502562523,
      "num_tokens": 48840515.0,
      "step": 26545
    },
    {
      "entropy": 4.960235738754273,
      "epoch": 2.217674574006014,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045050671650608513,
      "loss": 4.748,
      "mean_token_accuracy": 0.22058119177818297,
      "num_tokens": 48849248.0,
      "step": 26550
    },
    {
      "entropy": 5.085628843307495,
      "epoch": 2.2180922151687272,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004504880866355587,
      "loss": 4.7563,
      "mean_token_accuracy": 0.2184805914759636,
      "num_tokens": 48858593.0,
      "step": 26555
    },
    {
      "entropy": 5.101674318313599,
      "epoch": 2.21850985633144,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000450469453692902,
      "loss": 4.8015,
      "mean_token_accuracy": 0.21495840698480606,
      "num_tokens": 48867859.0,
      "step": 26560
    },
    {
      "entropy": 5.063493251800537,
      "epoch": 2.218927497494153,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045045081767844137,
      "loss": 4.747,
      "mean_token_accuracy": 0.23326435387134553,
      "num_tokens": 48877556.0,
      "step": 26565
    },
    {
      "entropy": 5.138156080245972,
      "epoch": 2.219345138656866,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000450432178592503,
      "loss": 4.7699,
      "mean_token_accuracy": 0.22375386357307434,
      "num_tokens": 48887468.0,
      "step": 26570
    },
    {
      "entropy": 5.029049682617187,
      "epoch": 2.219762779819579,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004504135364354131,
      "loss": 4.6802,
      "mean_token_accuracy": 0.2323159322142601,
      "num_tokens": 48895683.0,
      "step": 26575
    },
    {
      "entropy": 5.049517965316772,
      "epoch": 2.220180420982292,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045039489120749816,
      "loss": 4.8213,
      "mean_token_accuracy": 0.22113531231880187,
      "num_tokens": 48904522.0,
      "step": 26580
    },
    {
      "entropy": 5.128305435180664,
      "epoch": 2.220598062145005,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004503762429090845,
      "loss": 4.8621,
      "mean_token_accuracy": 0.2136386901140213,
      "num_tokens": 48914091.0,
      "step": 26585
    },
    {
      "entropy": 5.05969204902649,
      "epoch": 2.221015703307718,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004503575915404986,
      "loss": 4.7209,
      "mean_token_accuracy": 0.2222352772951126,
      "num_tokens": 48922652.0,
      "step": 26590
    },
    {
      "entropy": 5.090302467346191,
      "epoch": 2.221433344470431,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000450338937102067,
      "loss": 4.8626,
      "mean_token_accuracy": 0.2141646757721901,
      "num_tokens": 48931512.0,
      "step": 26595
    },
    {
      "entropy": 5.063072252273559,
      "epoch": 2.221850985633144,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045032027959411624,
      "loss": 4.7538,
      "mean_token_accuracy": 0.21398120522499084,
      "num_tokens": 48940456.0,
      "step": 26600
    },
    {
      "entropy": 5.096351289749146,
      "epoch": 2.222268626795857,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000450301619016973,
      "loss": 4.8803,
      "mean_token_accuracy": 0.21185212582349777,
      "num_tokens": 48951109.0,
      "step": 26605
    },
    {
      "entropy": 5.095682096481323,
      "epoch": 2.22268626795857,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045028295537096385,
      "loss": 4.6965,
      "mean_token_accuracy": 0.22654127776622773,
      "num_tokens": 48959586.0,
      "step": 26610
    },
    {
      "entropy": 5.010228252410888,
      "epoch": 2.223103909121283,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045026428865641563,
      "loss": 4.7327,
      "mean_token_accuracy": 0.22135053426027299,
      "num_tokens": 48968393.0,
      "step": 26615
    },
    {
      "entropy": 5.06455807685852,
      "epoch": 2.223521550283996,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000450245618873655,
      "loss": 4.7731,
      "mean_token_accuracy": 0.2161208540201187,
      "num_tokens": 48977483.0,
      "step": 26620
    },
    {
      "entropy": 5.141281318664551,
      "epoch": 2.223939191446709,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004502269460230089,
      "loss": 4.8146,
      "mean_token_accuracy": 0.21588003486394883,
      "num_tokens": 48986568.0,
      "step": 26625
    },
    {
      "entropy": 5.04870457649231,
      "epoch": 2.224356832609422,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004502082701048042,
      "loss": 4.7551,
      "mean_token_accuracy": 0.22036666572093963,
      "num_tokens": 48995952.0,
      "step": 26630
    },
    {
      "entropy": 5.056284999847412,
      "epoch": 2.224774473772135,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045018959111936785,
      "loss": 4.8264,
      "mean_token_accuracy": 0.21163361072540282,
      "num_tokens": 49006168.0,
      "step": 26635
    },
    {
      "entropy": 5.080404901504517,
      "epoch": 2.225192114934848,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004501709090670268,
      "loss": 4.711,
      "mean_token_accuracy": 0.22793580293655397,
      "num_tokens": 49014903.0,
      "step": 26640
    },
    {
      "entropy": 5.032731342315674,
      "epoch": 2.225609756097561,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004501522239481081,
      "loss": 4.7472,
      "mean_token_accuracy": 0.22681637108325958,
      "num_tokens": 49024801.0,
      "step": 26645
    },
    {
      "entropy": 5.073336219787597,
      "epoch": 2.2260273972602738,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045013353576293897,
      "loss": 4.736,
      "mean_token_accuracy": 0.22158463895320893,
      "num_tokens": 49033822.0,
      "step": 26650
    },
    {
      "entropy": 5.096467590332031,
      "epoch": 2.226445038422987,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045011484451184646,
      "loss": 4.7369,
      "mean_token_accuracy": 0.22521404922008514,
      "num_tokens": 49044074.0,
      "step": 26655
    },
    {
      "entropy": 5.016061639785766,
      "epoch": 2.2268626795857,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045009615019515783,
      "loss": 4.7565,
      "mean_token_accuracy": 0.22037094831466675,
      "num_tokens": 49053090.0,
      "step": 26660
    },
    {
      "entropy": 4.94171667098999,
      "epoch": 2.227280320748413,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045007745281320035,
      "loss": 4.7017,
      "mean_token_accuracy": 0.225450299680233,
      "num_tokens": 49061652.0,
      "step": 26665
    },
    {
      "entropy": 5.024256086349487,
      "epoch": 2.227697961911126,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045005875236630124,
      "loss": 4.6412,
      "mean_token_accuracy": 0.23165425807237625,
      "num_tokens": 49070196.0,
      "step": 26670
    },
    {
      "entropy": 5.0489091873168945,
      "epoch": 2.228115603073839,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045004004885478797,
      "loss": 4.7322,
      "mean_token_accuracy": 0.22304946929216385,
      "num_tokens": 49079422.0,
      "step": 26675
    },
    {
      "entropy": 4.985947751998902,
      "epoch": 2.228533244236552,
      "grad_norm": 1.21875,
      "learning_rate": 0.000450021342278988,
      "loss": 4.7374,
      "mean_token_accuracy": 0.21667269766330718,
      "num_tokens": 49089010.0,
      "step": 26680
    },
    {
      "entropy": 5.0174257278442385,
      "epoch": 2.228950885399265,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004500026326392287,
      "loss": 4.6658,
      "mean_token_accuracy": 0.2305363416671753,
      "num_tokens": 49097153.0,
      "step": 26685
    },
    {
      "entropy": 5.068580770492554,
      "epoch": 2.229368526561978,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044998391993583773,
      "loss": 4.7455,
      "mean_token_accuracy": 0.22254987061023712,
      "num_tokens": 49107741.0,
      "step": 26690
    },
    {
      "entropy": 5.052727365493775,
      "epoch": 2.229786167724691,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004499652041691425,
      "loss": 4.7625,
      "mean_token_accuracy": 0.22191012799739837,
      "num_tokens": 49117182.0,
      "step": 26695
    },
    {
      "entropy": 5.095395851135254,
      "epoch": 2.230203808887404,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004499464853394709,
      "loss": 4.8231,
      "mean_token_accuracy": 0.22072781026363372,
      "num_tokens": 49128036.0,
      "step": 26700
    },
    {
      "entropy": 5.069540977478027,
      "epoch": 2.230621450050117,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044992776344715036,
      "loss": 4.7212,
      "mean_token_accuracy": 0.2266220137476921,
      "num_tokens": 49137669.0,
      "step": 26705
    },
    {
      "entropy": 5.023878669738769,
      "epoch": 2.23103909121283,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004499090384925088,
      "loss": 4.7483,
      "mean_token_accuracy": 0.2200025200843811,
      "num_tokens": 49147447.0,
      "step": 26710
    },
    {
      "entropy": 5.090569877624512,
      "epoch": 2.231456732375543,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044989031047587406,
      "loss": 4.814,
      "mean_token_accuracy": 0.22096264064311982,
      "num_tokens": 49156516.0,
      "step": 26715
    },
    {
      "entropy": 5.099348354339599,
      "epoch": 2.231874373538256,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004498715793975738,
      "loss": 4.8381,
      "mean_token_accuracy": 0.21600982248783113,
      "num_tokens": 49165735.0,
      "step": 26720
    },
    {
      "entropy": 5.051617431640625,
      "epoch": 2.232292014700969,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044985284525793617,
      "loss": 4.7213,
      "mean_token_accuracy": 0.2225475564599037,
      "num_tokens": 49174389.0,
      "step": 26725
    },
    {
      "entropy": 5.024892997741699,
      "epoch": 2.232709655863682,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004498341080572889,
      "loss": 4.7128,
      "mean_token_accuracy": 0.2227322682738304,
      "num_tokens": 49183521.0,
      "step": 26730
    },
    {
      "entropy": 5.014713335037231,
      "epoch": 2.233127297026395,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044981536779596025,
      "loss": 4.6949,
      "mean_token_accuracy": 0.23190945833921434,
      "num_tokens": 49192971.0,
      "step": 26735
    },
    {
      "entropy": 5.054001188278198,
      "epoch": 2.233544938189108,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004497966244742781,
      "loss": 4.7084,
      "mean_token_accuracy": 0.2272268384695053,
      "num_tokens": 49201502.0,
      "step": 26740
    },
    {
      "entropy": 4.992499780654907,
      "epoch": 2.2339625793518207,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004497778780925706,
      "loss": 4.6857,
      "mean_token_accuracy": 0.22646575123071672,
      "num_tokens": 49211096.0,
      "step": 26745
    },
    {
      "entropy": 5.075872802734375,
      "epoch": 2.234380220514534,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044975912865116604,
      "loss": 4.8261,
      "mean_token_accuracy": 0.2176129013299942,
      "num_tokens": 49221447.0,
      "step": 26750
    },
    {
      "entropy": 5.072015428543091,
      "epoch": 2.234797861677247,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044974037615039253,
      "loss": 4.7099,
      "mean_token_accuracy": 0.2231420412659645,
      "num_tokens": 49229572.0,
      "step": 26755
    },
    {
      "entropy": 5.089960765838623,
      "epoch": 2.23521550283996,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044972162059057837,
      "loss": 4.7424,
      "mean_token_accuracy": 0.21762142330408096,
      "num_tokens": 49239184.0,
      "step": 26760
    },
    {
      "entropy": 5.029062366485595,
      "epoch": 2.235633144002673,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044970286197205196,
      "loss": 4.7483,
      "mean_token_accuracy": 0.21998733133077622,
      "num_tokens": 49248844.0,
      "step": 26765
    },
    {
      "entropy": 5.077652549743652,
      "epoch": 2.2360507851653857,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004496841002951416,
      "loss": 4.7182,
      "mean_token_accuracy": 0.22065978795289992,
      "num_tokens": 49258028.0,
      "step": 26770
    },
    {
      "entropy": 5.041831302642822,
      "epoch": 2.236468426328099,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004496653355601759,
      "loss": 4.7405,
      "mean_token_accuracy": 0.2216854065656662,
      "num_tokens": 49267590.0,
      "step": 26775
    },
    {
      "entropy": 5.036361169815064,
      "epoch": 2.2368860674908118,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004496465677674832,
      "loss": 4.6685,
      "mean_token_accuracy": 0.22826511263847352,
      "num_tokens": 49275897.0,
      "step": 26780
    },
    {
      "entropy": 5.037681531906128,
      "epoch": 2.237303708653525,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004496277969173921,
      "loss": 4.7547,
      "mean_token_accuracy": 0.21800228506326674,
      "num_tokens": 49285853.0,
      "step": 26785
    },
    {
      "entropy": 5.05037112236023,
      "epoch": 2.237721349816238,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044960902301023126,
      "loss": 4.6741,
      "mean_token_accuracy": 0.22566450238227845,
      "num_tokens": 49295129.0,
      "step": 26790
    },
    {
      "entropy": 4.912579202651978,
      "epoch": 2.238138990978951,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004495902460463293,
      "loss": 4.6006,
      "mean_token_accuracy": 0.22939294427633286,
      "num_tokens": 49304117.0,
      "step": 26795
    },
    {
      "entropy": 4.9903342723846436,
      "epoch": 2.238556632141664,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004495714660260149,
      "loss": 4.6183,
      "mean_token_accuracy": 0.22863979637622833,
      "num_tokens": 49312872.0,
      "step": 26800
    },
    {
      "entropy": 5.143255090713501,
      "epoch": 2.2389742733043767,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044955268294961687,
      "loss": 4.8631,
      "mean_token_accuracy": 0.21061184257268906,
      "num_tokens": 49322081.0,
      "step": 26805
    },
    {
      "entropy": 5.011145162582397,
      "epoch": 2.23939191446709,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044953389681746405,
      "loss": 4.6581,
      "mean_token_accuracy": 0.2254341274499893,
      "num_tokens": 49331288.0,
      "step": 26810
    },
    {
      "entropy": 4.962149953842163,
      "epoch": 2.2398095556298028,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004495151076298852,
      "loss": 4.7824,
      "mean_token_accuracy": 0.2198369711637497,
      "num_tokens": 49341135.0,
      "step": 26815
    },
    {
      "entropy": 5.080115222930909,
      "epoch": 2.240227196792516,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004494963153872093,
      "loss": 4.8134,
      "mean_token_accuracy": 0.22006449103355408,
      "num_tokens": 49350450.0,
      "step": 26820
    },
    {
      "entropy": 5.1306342601776125,
      "epoch": 2.240644837955229,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044947752008976546,
      "loss": 4.7848,
      "mean_token_accuracy": 0.2249903991818428,
      "num_tokens": 49358719.0,
      "step": 26825
    },
    {
      "entropy": 5.029606103897095,
      "epoch": 2.241062479117942,
      "grad_norm": 1.125,
      "learning_rate": 0.0004494587217378826,
      "loss": 4.6901,
      "mean_token_accuracy": 0.22959668636322023,
      "num_tokens": 49366976.0,
      "step": 26830
    },
    {
      "entropy": 4.985992527008056,
      "epoch": 2.241480120280655,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00044943992033188974,
      "loss": 4.7626,
      "mean_token_accuracy": 0.22284507751464844,
      "num_tokens": 49376256.0,
      "step": 26835
    },
    {
      "entropy": 4.992870712280274,
      "epoch": 2.2418977614433677,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004494211158721162,
      "loss": 4.6792,
      "mean_token_accuracy": 0.22882577180862426,
      "num_tokens": 49385120.0,
      "step": 26840
    },
    {
      "entropy": 5.102829265594482,
      "epoch": 2.242315402606081,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000449402308358891,
      "loss": 4.7632,
      "mean_token_accuracy": 0.21531153321266175,
      "num_tokens": 49394960.0,
      "step": 26845
    },
    {
      "entropy": 4.943449258804321,
      "epoch": 2.242733043768794,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004493834977925435,
      "loss": 4.6214,
      "mean_token_accuracy": 0.23612059503793717,
      "num_tokens": 49404332.0,
      "step": 26850
    },
    {
      "entropy": 5.002291250228882,
      "epoch": 2.243150684931507,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044936468417340294,
      "loss": 4.6867,
      "mean_token_accuracy": 0.22386109977960586,
      "num_tokens": 49413723.0,
      "step": 26855
    },
    {
      "entropy": 5.038396787643433,
      "epoch": 2.24356832609422,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004493458675017987,
      "loss": 4.7783,
      "mean_token_accuracy": 0.22020470201969147,
      "num_tokens": 49423188.0,
      "step": 26860
    },
    {
      "entropy": 5.0254271030426025,
      "epoch": 2.2439859672569327,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004493270477780602,
      "loss": 4.7657,
      "mean_token_accuracy": 0.22333473116159439,
      "num_tokens": 49432826.0,
      "step": 26865
    },
    {
      "entropy": 5.068181753158569,
      "epoch": 2.244403608419646,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044930822500251693,
      "loss": 4.7385,
      "mean_token_accuracy": 0.2209090232849121,
      "num_tokens": 49442538.0,
      "step": 26870
    },
    {
      "entropy": 5.107479810714722,
      "epoch": 2.2448212495823587,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044928939917549833,
      "loss": 4.7908,
      "mean_token_accuracy": 0.22086121886968613,
      "num_tokens": 49451643.0,
      "step": 26875
    },
    {
      "entropy": 5.100422763824463,
      "epoch": 2.245238890745072,
      "grad_norm": 1.09375,
      "learning_rate": 0.000449270570297334,
      "loss": 4.7616,
      "mean_token_accuracy": 0.22636270821094512,
      "num_tokens": 49460866.0,
      "step": 26880
    },
    {
      "entropy": 5.055569267272949,
      "epoch": 2.245656531907785,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004492517383683536,
      "loss": 4.7615,
      "mean_token_accuracy": 0.21379673928022386,
      "num_tokens": 49470750.0,
      "step": 26885
    },
    {
      "entropy": 4.980667066574097,
      "epoch": 2.2460741730704976,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044923290338888674,
      "loss": 4.7517,
      "mean_token_accuracy": 0.2230258420109749,
      "num_tokens": 49479841.0,
      "step": 26890
    },
    {
      "entropy": 4.9946308612823485,
      "epoch": 2.246491814233211,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004492140653592632,
      "loss": 4.7433,
      "mean_token_accuracy": 0.21860391646623611,
      "num_tokens": 49488877.0,
      "step": 26895
    },
    {
      "entropy": 5.069811868667602,
      "epoch": 2.2469094553959237,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044919522427981276,
      "loss": 4.8093,
      "mean_token_accuracy": 0.21863721013069154,
      "num_tokens": 49498892.0,
      "step": 26900
    },
    {
      "entropy": 5.13899712562561,
      "epoch": 2.247327096558637,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004491763801508652,
      "loss": 4.84,
      "mean_token_accuracy": 0.21099752187728882,
      "num_tokens": 49507444.0,
      "step": 26905
    },
    {
      "entropy": 5.0231091499328615,
      "epoch": 2.2477447377213498,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044915753297275046,
      "loss": 4.7267,
      "mean_token_accuracy": 0.2289866864681244,
      "num_tokens": 49516710.0,
      "step": 26910
    },
    {
      "entropy": 5.057309484481811,
      "epoch": 2.248162378884063,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00044913868274579856,
      "loss": 4.7711,
      "mean_token_accuracy": 0.22655697166919708,
      "num_tokens": 49526945.0,
      "step": 26915
    },
    {
      "entropy": 4.964813327789306,
      "epoch": 2.248580020046776,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044911982947033925,
      "loss": 4.6339,
      "mean_token_accuracy": 0.22779639512300492,
      "num_tokens": 49536058.0,
      "step": 26920
    },
    {
      "entropy": 4.996617650985717,
      "epoch": 2.2489976612094886,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004491009731467028,
      "loss": 4.7132,
      "mean_token_accuracy": 0.22560816258192062,
      "num_tokens": 49544761.0,
      "step": 26925
    },
    {
      "entropy": 5.190349912643432,
      "epoch": 2.249415302372202,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004490821137752193,
      "loss": 4.8675,
      "mean_token_accuracy": 0.20670540034770965,
      "num_tokens": 49553933.0,
      "step": 26930
    },
    {
      "entropy": 5.073377990722657,
      "epoch": 2.2498329435349147,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004490632513562188,
      "loss": 4.7251,
      "mean_token_accuracy": 0.22336397916078568,
      "num_tokens": 49563668.0,
      "step": 26935
    },
    {
      "entropy": 5.086310243606567,
      "epoch": 2.250250584697628,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004490443858900316,
      "loss": 4.7747,
      "mean_token_accuracy": 0.22098649442195892,
      "num_tokens": 49573830.0,
      "step": 26940
    },
    {
      "entropy": 4.975074291229248,
      "epoch": 2.2506682258603408,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044902551737698785,
      "loss": 4.7076,
      "mean_token_accuracy": 0.22971808165311813,
      "num_tokens": 49582396.0,
      "step": 26945
    },
    {
      "entropy": 5.030947923660278,
      "epoch": 2.251085867023054,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044900664581741805,
      "loss": 4.6961,
      "mean_token_accuracy": 0.22673072963953017,
      "num_tokens": 49590918.0,
      "step": 26950
    },
    {
      "entropy": 5.046582221984863,
      "epoch": 2.251503508185767,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004489877712116524,
      "loss": 4.7685,
      "mean_token_accuracy": 0.2171344429254532,
      "num_tokens": 49600365.0,
      "step": 26955
    },
    {
      "entropy": 5.047661542892456,
      "epoch": 2.2519211493484796,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004489688935600214,
      "loss": 4.7726,
      "mean_token_accuracy": 0.2258819669485092,
      "num_tokens": 49610002.0,
      "step": 26960
    },
    {
      "entropy": 5.035359954833984,
      "epoch": 2.252338790511193,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044895001286285557,
      "loss": 4.7232,
      "mean_token_accuracy": 0.2275412604212761,
      "num_tokens": 49619085.0,
      "step": 26965
    },
    {
      "entropy": 4.945445156097412,
      "epoch": 2.2527564316739057,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044893112912048526,
      "loss": 4.5936,
      "mean_token_accuracy": 0.23313352167606355,
      "num_tokens": 49627753.0,
      "step": 26970
    },
    {
      "entropy": 5.002104425430298,
      "epoch": 2.253174072836619,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004489122423332413,
      "loss": 4.6801,
      "mean_token_accuracy": 0.22654951810836793,
      "num_tokens": 49637377.0,
      "step": 26975
    },
    {
      "entropy": 5.01311240196228,
      "epoch": 2.253591713999332,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044889335250145415,
      "loss": 4.7579,
      "mean_token_accuracy": 0.21827410906553268,
      "num_tokens": 49645323.0,
      "step": 26980
    },
    {
      "entropy": 5.073163366317749,
      "epoch": 2.2540093551620446,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004488744596254545,
      "loss": 4.7538,
      "mean_token_accuracy": 0.2202336460351944,
      "num_tokens": 49655076.0,
      "step": 26985
    },
    {
      "entropy": 5.029118299484253,
      "epoch": 2.254426996324758,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044885556370557317,
      "loss": 4.7784,
      "mean_token_accuracy": 0.22609878778457643,
      "num_tokens": 49664518.0,
      "step": 26990
    },
    {
      "entropy": 4.944000148773194,
      "epoch": 2.2548446374874707,
      "grad_norm": 1.109375,
      "learning_rate": 0.000448836664742141,
      "loss": 4.5628,
      "mean_token_accuracy": 0.23488419353961945,
      "num_tokens": 49673530.0,
      "step": 26995
    },
    {
      "entropy": 5.106258153915405,
      "epoch": 2.255262278650184,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004488177627354887,
      "loss": 4.7987,
      "mean_token_accuracy": 0.21408420652151108,
      "num_tokens": 49682716.0,
      "step": 27000
    },
    {
      "epoch": 2.255262278650184,
      "eval_entropy": 4.8098025078037185,
      "eval_loss": 4.802053451538086,
      "eval_mean_token_accuracy": 0.23057828838780153,
      "eval_num_tokens": 49682716.0,
      "eval_runtime": 27.6347,
      "eval_samples_per_second": 1366.507,
      "eval_steps_per_second": 170.836,
      "step": 27000
    },
    {
      "entropy": 5.0329506397247314,
      "epoch": 2.2556799198128967,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044879885768594726,
      "loss": 4.6874,
      "mean_token_accuracy": 0.22076817750930786,
      "num_tokens": 49691726.0,
      "step": 27005
    },
    {
      "entropy": 4.905385494232178,
      "epoch": 2.2560975609756095,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044877994959384757,
      "loss": 4.6495,
      "mean_token_accuracy": 0.22539964765310289,
      "num_tokens": 49701021.0,
      "step": 27010
    },
    {
      "entropy": 5.004996013641358,
      "epoch": 2.256515202138323,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004487610384595207,
      "loss": 4.7131,
      "mean_token_accuracy": 0.22081806063652037,
      "num_tokens": 49710583.0,
      "step": 27015
    },
    {
      "entropy": 5.054051685333252,
      "epoch": 2.2569328433010356,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004487421242832977,
      "loss": 4.7793,
      "mean_token_accuracy": 0.2258690044283867,
      "num_tokens": 49718365.0,
      "step": 27020
    },
    {
      "entropy": 4.996800947189331,
      "epoch": 2.257350484463749,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004487232070655097,
      "loss": 4.7065,
      "mean_token_accuracy": 0.2272746443748474,
      "num_tokens": 49728340.0,
      "step": 27025
    },
    {
      "entropy": 4.949030494689941,
      "epoch": 2.2577681256264617,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004487042868064878,
      "loss": 4.581,
      "mean_token_accuracy": 0.2353675678372383,
      "num_tokens": 49736061.0,
      "step": 27030
    },
    {
      "entropy": 4.979730796813965,
      "epoch": 2.258185766789175,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004486853635065634,
      "loss": 4.6313,
      "mean_token_accuracy": 0.2341872051358223,
      "num_tokens": 49744797.0,
      "step": 27035
    },
    {
      "entropy": 5.04857816696167,
      "epoch": 2.2586034079518877,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044866643716606747,
      "loss": 4.7153,
      "mean_token_accuracy": 0.22977166324853898,
      "num_tokens": 49754118.0,
      "step": 27040
    },
    {
      "entropy": 5.165176677703857,
      "epoch": 2.2590210491146006,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004486475077853316,
      "loss": 4.935,
      "mean_token_accuracy": 0.20594364106655122,
      "num_tokens": 49764721.0,
      "step": 27045
    },
    {
      "entropy": 5.100328397750855,
      "epoch": 2.259438690277314,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00044862857536468717,
      "loss": 4.7283,
      "mean_token_accuracy": 0.22435138523578643,
      "num_tokens": 49773013.0,
      "step": 27050
    },
    {
      "entropy": 5.084353494644165,
      "epoch": 2.2598563314400266,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004486096399044654,
      "loss": 4.8502,
      "mean_token_accuracy": 0.21880890429019928,
      "num_tokens": 49781608.0,
      "step": 27055
    },
    {
      "entropy": 5.169388961791992,
      "epoch": 2.26027397260274,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000448590701404998,
      "loss": 4.7978,
      "mean_token_accuracy": 0.2177301228046417,
      "num_tokens": 49791136.0,
      "step": 27060
    },
    {
      "entropy": 5.025900268554688,
      "epoch": 2.2606916137654527,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044857175986661634,
      "loss": 4.6645,
      "mean_token_accuracy": 0.22739135026931762,
      "num_tokens": 49800656.0,
      "step": 27065
    },
    {
      "entropy": 5.079333925247193,
      "epoch": 2.261109254928166,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004485528152896522,
      "loss": 4.8749,
      "mean_token_accuracy": 0.20850251913070678,
      "num_tokens": 49810528.0,
      "step": 27070
    },
    {
      "entropy": 5.042648077011108,
      "epoch": 2.2615268960908788,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044853386767443706,
      "loss": 4.7228,
      "mean_token_accuracy": 0.22198313027620314,
      "num_tokens": 49819979.0,
      "step": 27075
    },
    {
      "entropy": 5.088203716278076,
      "epoch": 2.2619445372535916,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044851491702130265,
      "loss": 4.761,
      "mean_token_accuracy": 0.2231805592775345,
      "num_tokens": 49828656.0,
      "step": 27080
    },
    {
      "entropy": 5.085560703277588,
      "epoch": 2.262362178416305,
      "grad_norm": 1.265625,
      "learning_rate": 0.00044849596333058087,
      "loss": 4.8225,
      "mean_token_accuracy": 0.21462855637073516,
      "num_tokens": 49838103.0,
      "step": 27085
    },
    {
      "entropy": 5.113306665420533,
      "epoch": 2.2627798195790176,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044847700660260325,
      "loss": 4.8175,
      "mean_token_accuracy": 0.21545398980379105,
      "num_tokens": 49847863.0,
      "step": 27090
    },
    {
      "entropy": 5.040954542160034,
      "epoch": 2.263197460741731,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004484580468377019,
      "loss": 4.7285,
      "mean_token_accuracy": 0.22286092042922973,
      "num_tokens": 49857608.0,
      "step": 27095
    },
    {
      "entropy": 4.983330345153808,
      "epoch": 2.2636151019044437,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004484390840362086,
      "loss": 4.6633,
      "mean_token_accuracy": 0.2233608439564705,
      "num_tokens": 49867805.0,
      "step": 27100
    },
    {
      "entropy": 5.071676778793335,
      "epoch": 2.2640327430671565,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044842011819845546,
      "loss": 4.821,
      "mean_token_accuracy": 0.21873106807470322,
      "num_tokens": 49877456.0,
      "step": 27105
    },
    {
      "entropy": 5.141354894638061,
      "epoch": 2.2644503842298698,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044840114932477435,
      "loss": 4.7219,
      "mean_token_accuracy": 0.22108578383922578,
      "num_tokens": 49886179.0,
      "step": 27110
    },
    {
      "entropy": 5.13409571647644,
      "epoch": 2.2648680253925826,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004483821774154975,
      "loss": 4.8155,
      "mean_token_accuracy": 0.2177382379770279,
      "num_tokens": 49895765.0,
      "step": 27115
    },
    {
      "entropy": 5.028174924850464,
      "epoch": 2.265285666555296,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044836320247095674,
      "loss": 4.8097,
      "mean_token_accuracy": 0.2132649838924408,
      "num_tokens": 49904921.0,
      "step": 27120
    },
    {
      "entropy": 5.059199047088623,
      "epoch": 2.2657033077180087,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004483442244914846,
      "loss": 4.7683,
      "mean_token_accuracy": 0.2237204357981682,
      "num_tokens": 49913464.0,
      "step": 27125
    },
    {
      "entropy": 5.157733726501465,
      "epoch": 2.2661209488807215,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004483252434774131,
      "loss": 4.8163,
      "mean_token_accuracy": 0.22076777517795562,
      "num_tokens": 49922893.0,
      "step": 27130
    },
    {
      "entropy": 5.025922060012817,
      "epoch": 2.2665385900434347,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004483062594290746,
      "loss": 4.7126,
      "mean_token_accuracy": 0.2298695370554924,
      "num_tokens": 49932011.0,
      "step": 27135
    },
    {
      "entropy": 5.101520824432373,
      "epoch": 2.2669562312061475,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004482872723468014,
      "loss": 4.8717,
      "mean_token_accuracy": 0.21071176528930663,
      "num_tokens": 49940971.0,
      "step": 27140
    },
    {
      "entropy": 5.137262296676636,
      "epoch": 2.267373872368861,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00044826828223092584,
      "loss": 4.7905,
      "mean_token_accuracy": 0.21976349353790284,
      "num_tokens": 49951533.0,
      "step": 27145
    },
    {
      "entropy": 5.096424674987793,
      "epoch": 2.2677915135315736,
      "grad_norm": 1.125,
      "learning_rate": 0.0004482492890817806,
      "loss": 4.7536,
      "mean_token_accuracy": 0.22541194707155227,
      "num_tokens": 49960842.0,
      "step": 27150
    },
    {
      "entropy": 5.090888929367066,
      "epoch": 2.268209154694287,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044823029289969786,
      "loss": 4.8043,
      "mean_token_accuracy": 0.22075025290250777,
      "num_tokens": 49970814.0,
      "step": 27155
    },
    {
      "entropy": 5.106673955917358,
      "epoch": 2.2686267958569997,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004482112936850104,
      "loss": 4.7811,
      "mean_token_accuracy": 0.22353345155715942,
      "num_tokens": 49980912.0,
      "step": 27160
    },
    {
      "entropy": 5.075198841094971,
      "epoch": 2.2690444370197125,
      "grad_norm": 1.3125,
      "learning_rate": 0.00044819229143805076,
      "loss": 4.8322,
      "mean_token_accuracy": 0.22340212613344193,
      "num_tokens": 49990404.0,
      "step": 27165
    },
    {
      "entropy": 4.969185495376587,
      "epoch": 2.2694620781824257,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004481732861591516,
      "loss": 4.6412,
      "mean_token_accuracy": 0.2372678339481354,
      "num_tokens": 50000569.0,
      "step": 27170
    },
    {
      "entropy": 5.009132528305054,
      "epoch": 2.2698797193451385,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004481542778486456,
      "loss": 4.7613,
      "mean_token_accuracy": 0.22208941131830215,
      "num_tokens": 50009332.0,
      "step": 27175
    },
    {
      "entropy": 5.089830350875855,
      "epoch": 2.270297360507852,
      "grad_norm": 1.125,
      "learning_rate": 0.0004481352665068654,
      "loss": 4.7678,
      "mean_token_accuracy": 0.21631626039743423,
      "num_tokens": 50019184.0,
      "step": 27180
    },
    {
      "entropy": 5.152768945693969,
      "epoch": 2.2707150016705646,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044811625213414416,
      "loss": 4.8987,
      "mean_token_accuracy": 0.21099889129400254,
      "num_tokens": 50028522.0,
      "step": 27185
    },
    {
      "entropy": 5.151764822006226,
      "epoch": 2.271132642833278,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004480972347308145,
      "loss": 4.7856,
      "mean_token_accuracy": 0.21712990552186967,
      "num_tokens": 50037850.0,
      "step": 27190
    },
    {
      "entropy": 4.965018081665039,
      "epoch": 2.2715502839959907,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004480782142972093,
      "loss": 4.7042,
      "mean_token_accuracy": 0.2243673175573349,
      "num_tokens": 50047096.0,
      "step": 27195
    },
    {
      "entropy": 5.079360008239746,
      "epoch": 2.2719679251587035,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044805919083366175,
      "loss": 4.7133,
      "mean_token_accuracy": 0.2201509118080139,
      "num_tokens": 50055847.0,
      "step": 27200
    },
    {
      "entropy": 5.104928827285766,
      "epoch": 2.2723855663214167,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00044804016434050466,
      "loss": 4.9419,
      "mean_token_accuracy": 0.20714229196310044,
      "num_tokens": 50064498.0,
      "step": 27205
    },
    {
      "entropy": 5.076524782180786,
      "epoch": 2.2728032074841296,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004480211348180712,
      "loss": 4.7978,
      "mean_token_accuracy": 0.22680406123399735,
      "num_tokens": 50072831.0,
      "step": 27210
    },
    {
      "entropy": 5.157875061035156,
      "epoch": 2.273220848646843,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044800210226669454,
      "loss": 4.7924,
      "mean_token_accuracy": 0.22358013242483138,
      "num_tokens": 50082559.0,
      "step": 27215
    },
    {
      "entropy": 5.032923030853271,
      "epoch": 2.2736384898095556,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004479830666867078,
      "loss": 4.7464,
      "mean_token_accuracy": 0.22562611997127532,
      "num_tokens": 50091853.0,
      "step": 27220
    },
    {
      "entropy": 5.037526369094849,
      "epoch": 2.2740561309722684,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004479640280784443,
      "loss": 4.717,
      "mean_token_accuracy": 0.22317264676094056,
      "num_tokens": 50100601.0,
      "step": 27225
    },
    {
      "entropy": 5.051538419723511,
      "epoch": 2.2744737721349817,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004479449864422373,
      "loss": 4.7467,
      "mean_token_accuracy": 0.2226976901292801,
      "num_tokens": 50108841.0,
      "step": 27230
    },
    {
      "entropy": 5.063524961471558,
      "epoch": 2.2748914132976945,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004479259417784201,
      "loss": 4.7311,
      "mean_token_accuracy": 0.21681490540504456,
      "num_tokens": 50117755.0,
      "step": 27235
    },
    {
      "entropy": 5.0570801258087155,
      "epoch": 2.2753090544604078,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044790689408732623,
      "loss": 4.6968,
      "mean_token_accuracy": 0.23333998322486876,
      "num_tokens": 50127164.0,
      "step": 27240
    },
    {
      "entropy": 5.069578790664673,
      "epoch": 2.2757266956231206,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044788784336928896,
      "loss": 4.7272,
      "mean_token_accuracy": 0.2270131379365921,
      "num_tokens": 50136537.0,
      "step": 27245
    },
    {
      "entropy": 5.071907234191895,
      "epoch": 2.2761443367858334,
      "grad_norm": 1.125,
      "learning_rate": 0.0004478687896246419,
      "loss": 4.7424,
      "mean_token_accuracy": 0.21462854146957397,
      "num_tokens": 50146037.0,
      "step": 27250
    },
    {
      "entropy": 5.107400178909302,
      "epoch": 2.2765619779485466,
      "grad_norm": 1.125,
      "learning_rate": 0.00044784973285371863,
      "loss": 4.7842,
      "mean_token_accuracy": 0.21904889047145842,
      "num_tokens": 50154607.0,
      "step": 27255
    },
    {
      "entropy": 5.081778860092163,
      "epoch": 2.2769796191112595,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044783067305685274,
      "loss": 4.7346,
      "mean_token_accuracy": 0.21972626298666,
      "num_tokens": 50163801.0,
      "step": 27260
    },
    {
      "entropy": 5.081685161590576,
      "epoch": 2.2773972602739727,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004478116102343778,
      "loss": 4.8386,
      "mean_token_accuracy": 0.21131048798561097,
      "num_tokens": 50174237.0,
      "step": 27265
    },
    {
      "entropy": 5.100523948669434,
      "epoch": 2.2778149014366855,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004477925443866277,
      "loss": 4.7579,
      "mean_token_accuracy": 0.22147203087806702,
      "num_tokens": 50184354.0,
      "step": 27270
    },
    {
      "entropy": 5.1484168529510494,
      "epoch": 2.278232542599399,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000447773475513936,
      "loss": 4.8004,
      "mean_token_accuracy": 0.21419070065021514,
      "num_tokens": 50194286.0,
      "step": 27275
    },
    {
      "entropy": 5.03337082862854,
      "epoch": 2.2786501837621116,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00044775440361663677,
      "loss": 4.7254,
      "mean_token_accuracy": 0.22563909888267517,
      "num_tokens": 50203631.0,
      "step": 27280
    },
    {
      "entropy": 5.080928707122803,
      "epoch": 2.2790678249248244,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044773532869506373,
      "loss": 4.7881,
      "mean_token_accuracy": 0.21741534620523453,
      "num_tokens": 50213509.0,
      "step": 27285
    },
    {
      "entropy": 5.114600419998169,
      "epoch": 2.2794854660875377,
      "grad_norm": 1.125,
      "learning_rate": 0.0004477162507495508,
      "loss": 4.811,
      "mean_token_accuracy": 0.21769158989191056,
      "num_tokens": 50222747.0,
      "step": 27290
    },
    {
      "entropy": 5.109318351745605,
      "epoch": 2.2799031072502505,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000447697169780432,
      "loss": 4.798,
      "mean_token_accuracy": 0.22063586115837097,
      "num_tokens": 50232711.0,
      "step": 27295
    },
    {
      "entropy": 4.964807510375977,
      "epoch": 2.2803207484129637,
      "grad_norm": 1.125,
      "learning_rate": 0.00044767808578804135,
      "loss": 4.592,
      "mean_token_accuracy": 0.23735909759998322,
      "num_tokens": 50241253.0,
      "step": 27300
    },
    {
      "entropy": 4.929737329483032,
      "epoch": 2.2807383895756765,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044765899877271306,
      "loss": 4.6239,
      "mean_token_accuracy": 0.22885268479585646,
      "num_tokens": 50249559.0,
      "step": 27305
    },
    {
      "entropy": 5.103724813461303,
      "epoch": 2.28115603073839,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044763990873478106,
      "loss": 4.836,
      "mean_token_accuracy": 0.2120375081896782,
      "num_tokens": 50259442.0,
      "step": 27310
    },
    {
      "entropy": 5.158670711517334,
      "epoch": 2.2815736719011026,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044762081567457966,
      "loss": 4.8073,
      "mean_token_accuracy": 0.22192182540893554,
      "num_tokens": 50268468.0,
      "step": 27315
    },
    {
      "entropy": 5.038602876663208,
      "epoch": 2.2819913130638154,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004476017195924432,
      "loss": 4.6722,
      "mean_token_accuracy": 0.2268402546644211,
      "num_tokens": 50278342.0,
      "step": 27320
    },
    {
      "entropy": 4.985432529449463,
      "epoch": 2.2824089542265287,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044758262048870574,
      "loss": 4.7008,
      "mean_token_accuracy": 0.22958674877882004,
      "num_tokens": 50287364.0,
      "step": 27325
    },
    {
      "entropy": 5.133501243591309,
      "epoch": 2.2828265953892415,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044756351836370184,
      "loss": 4.8418,
      "mean_token_accuracy": 0.21390993148088455,
      "num_tokens": 50297306.0,
      "step": 27330
    },
    {
      "entropy": 5.166635227203369,
      "epoch": 2.2832442365519547,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004475444132177659,
      "loss": 4.7339,
      "mean_token_accuracy": 0.22270148992538452,
      "num_tokens": 50306774.0,
      "step": 27335
    },
    {
      "entropy": 5.11018443107605,
      "epoch": 2.2836618777146676,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004475253050512322,
      "loss": 4.7377,
      "mean_token_accuracy": 0.2208909273147583,
      "num_tokens": 50316492.0,
      "step": 27340
    },
    {
      "entropy": 4.984866094589234,
      "epoch": 2.2840795188773804,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044750619386443543,
      "loss": 4.6871,
      "mean_token_accuracy": 0.22648324817419052,
      "num_tokens": 50324950.0,
      "step": 27345
    },
    {
      "entropy": 5.162129068374634,
      "epoch": 2.2844971600400936,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044748707965771005,
      "loss": 4.8668,
      "mean_token_accuracy": 0.2098850965499878,
      "num_tokens": 50334383.0,
      "step": 27350
    },
    {
      "entropy": 5.04634461402893,
      "epoch": 2.2849148012028064,
      "grad_norm": 1.125,
      "learning_rate": 0.00044746796243139077,
      "loss": 4.7711,
      "mean_token_accuracy": 0.22140702456235886,
      "num_tokens": 50343929.0,
      "step": 27355
    },
    {
      "entropy": 5.044144010543823,
      "epoch": 2.2853324423655197,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044744884218581217,
      "loss": 4.7486,
      "mean_token_accuracy": 0.21995137631893158,
      "num_tokens": 50352975.0,
      "step": 27360
    },
    {
      "entropy": 5.11721248626709,
      "epoch": 2.2857500835282325,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000447429718921309,
      "loss": 4.8129,
      "mean_token_accuracy": 0.2174692526459694,
      "num_tokens": 50362800.0,
      "step": 27365
    },
    {
      "entropy": 5.046429252624511,
      "epoch": 2.2861677246909453,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004474105926382161,
      "loss": 4.7635,
      "mean_token_accuracy": 0.22012146562337875,
      "num_tokens": 50372496.0,
      "step": 27370
    },
    {
      "entropy": 5.045615720748901,
      "epoch": 2.2865853658536586,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004473914633368683,
      "loss": 4.697,
      "mean_token_accuracy": 0.22658740282058715,
      "num_tokens": 50380634.0,
      "step": 27375
    },
    {
      "entropy": 5.029886293411255,
      "epoch": 2.2870030070163714,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004473723310176003,
      "loss": 4.6616,
      "mean_token_accuracy": 0.23139167129993438,
      "num_tokens": 50389318.0,
      "step": 27380
    },
    {
      "entropy": 5.0119733810424805,
      "epoch": 2.2874206481790846,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004473531956807471,
      "loss": 4.7632,
      "mean_token_accuracy": 0.22519397884607315,
      "num_tokens": 50398507.0,
      "step": 27385
    },
    {
      "entropy": 5.013142442703247,
      "epoch": 2.2878382893417974,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044733405732664394,
      "loss": 4.6325,
      "mean_token_accuracy": 0.22883139848709105,
      "num_tokens": 50406905.0,
      "step": 27390
    },
    {
      "entropy": 5.087789392471313,
      "epoch": 2.2882559305045107,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004473149159556255,
      "loss": 4.6919,
      "mean_token_accuracy": 0.22352370321750642,
      "num_tokens": 50415959.0,
      "step": 27395
    },
    {
      "entropy": 5.123177099227905,
      "epoch": 2.2886735716672235,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044729577156802714,
      "loss": 4.7827,
      "mean_token_accuracy": 0.21735214442014694,
      "num_tokens": 50425364.0,
      "step": 27400
    },
    {
      "entropy": 5.062253856658936,
      "epoch": 2.2890912128299363,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004472766241641839,
      "loss": 4.7761,
      "mean_token_accuracy": 0.22303968220949172,
      "num_tokens": 50434349.0,
      "step": 27405
    },
    {
      "entropy": 5.02071852684021,
      "epoch": 2.2895088539926496,
      "grad_norm": 1.171875,
      "learning_rate": 0.000447257473744431,
      "loss": 4.6703,
      "mean_token_accuracy": 0.22444139420986176,
      "num_tokens": 50442964.0,
      "step": 27410
    },
    {
      "entropy": 5.077487134933472,
      "epoch": 2.2899264951553624,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004472383203091036,
      "loss": 4.7871,
      "mean_token_accuracy": 0.216795851290226,
      "num_tokens": 50451625.0,
      "step": 27415
    },
    {
      "entropy": 5.00605583190918,
      "epoch": 2.2903441363180757,
      "grad_norm": 1.125,
      "learning_rate": 0.0004472191638585371,
      "loss": 4.6203,
      "mean_token_accuracy": 0.22907380014657974,
      "num_tokens": 50460062.0,
      "step": 27420
    },
    {
      "entropy": 4.983326721191406,
      "epoch": 2.2907617774807885,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044720000439306694,
      "loss": 4.634,
      "mean_token_accuracy": 0.22889379113912584,
      "num_tokens": 50468207.0,
      "step": 27425
    },
    {
      "entropy": 5.05954065322876,
      "epoch": 2.2911794186435017,
      "grad_norm": 1.234375,
      "learning_rate": 0.00044718084191302826,
      "loss": 4.749,
      "mean_token_accuracy": 0.2138363689184189,
      "num_tokens": 50478151.0,
      "step": 27430
    },
    {
      "entropy": 5.097065591812134,
      "epoch": 2.2915970598062145,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044716167641875685,
      "loss": 4.8017,
      "mean_token_accuracy": 0.21972308605909346,
      "num_tokens": 50487180.0,
      "step": 27435
    },
    {
      "entropy": 4.985680055618286,
      "epoch": 2.2920147009689273,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000447142507910588,
      "loss": 4.685,
      "mean_token_accuracy": 0.22252208292484282,
      "num_tokens": 50495661.0,
      "step": 27440
    },
    {
      "entropy": 5.070184373855591,
      "epoch": 2.2924323421316406,
      "grad_norm": 1.25,
      "learning_rate": 0.00044712333638885726,
      "loss": 4.8204,
      "mean_token_accuracy": 0.21602596193552018,
      "num_tokens": 50505854.0,
      "step": 27445
    },
    {
      "entropy": 5.1187999725341795,
      "epoch": 2.2928499832943534,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044710416185390043,
      "loss": 4.7298,
      "mean_token_accuracy": 0.22328036278486252,
      "num_tokens": 50514788.0,
      "step": 27450
    },
    {
      "entropy": 4.989723539352417,
      "epoch": 2.2932676244570667,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004470849843060531,
      "loss": 4.683,
      "mean_token_accuracy": 0.22649256736040116,
      "num_tokens": 50523537.0,
      "step": 27455
    },
    {
      "entropy": 5.057803153991699,
      "epoch": 2.2936852656197795,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044706580374565094,
      "loss": 4.7316,
      "mean_token_accuracy": 0.22108486890792847,
      "num_tokens": 50533165.0,
      "step": 27460
    },
    {
      "entropy": 5.090937757492066,
      "epoch": 2.2941029067824923,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004470466201730298,
      "loss": 4.7228,
      "mean_token_accuracy": 0.22410656362771988,
      "num_tokens": 50541531.0,
      "step": 27465
    },
    {
      "entropy": 5.061096858978272,
      "epoch": 2.2945205479452055,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004470274335885254,
      "loss": 4.6566,
      "mean_token_accuracy": 0.23031342327594756,
      "num_tokens": 50551097.0,
      "step": 27470
    },
    {
      "entropy": 5.021644926071167,
      "epoch": 2.2949381891079184,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004470082439924737,
      "loss": 4.7363,
      "mean_token_accuracy": 0.2259761407971382,
      "num_tokens": 50559796.0,
      "step": 27475
    },
    {
      "entropy": 5.033962059020996,
      "epoch": 2.2953558302706316,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004469890513852107,
      "loss": 4.8197,
      "mean_token_accuracy": 0.2118042454123497,
      "num_tokens": 50568671.0,
      "step": 27480
    },
    {
      "entropy": 5.115182018280029,
      "epoch": 2.2957734714333444,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004469698557670724,
      "loss": 4.7249,
      "mean_token_accuracy": 0.22385161817073823,
      "num_tokens": 50577396.0,
      "step": 27485
    },
    {
      "entropy": 5.099190616607666,
      "epoch": 2.2961911125960572,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004469506571383947,
      "loss": 4.743,
      "mean_token_accuracy": 0.22090187966823577,
      "num_tokens": 50587018.0,
      "step": 27490
    },
    {
      "entropy": 5.084864282608033,
      "epoch": 2.2966087537587705,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004469314554995137,
      "loss": 4.7991,
      "mean_token_accuracy": 0.217988058924675,
      "num_tokens": 50596026.0,
      "step": 27495
    },
    {
      "entropy": 5.004279327392578,
      "epoch": 2.2970263949214833,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044691225085076577,
      "loss": 4.6631,
      "mean_token_accuracy": 0.23329963088035582,
      "num_tokens": 50604973.0,
      "step": 27500
    },
    {
      "entropy": 5.070153045654297,
      "epoch": 2.2974440360841966,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004468930431924869,
      "loss": 4.7835,
      "mean_token_accuracy": 0.21965770274400712,
      "num_tokens": 50613686.0,
      "step": 27505
    },
    {
      "entropy": 5.089167833328247,
      "epoch": 2.2978616772469094,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044687383252501337,
      "loss": 4.8228,
      "mean_token_accuracy": 0.22393725514411927,
      "num_tokens": 50622840.0,
      "step": 27510
    },
    {
      "entropy": 4.9798613548278805,
      "epoch": 2.2982793184096226,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004468546188486815,
      "loss": 4.7214,
      "mean_token_accuracy": 0.2305450424551964,
      "num_tokens": 50631818.0,
      "step": 27515
    },
    {
      "entropy": 5.04114727973938,
      "epoch": 2.2986969595723354,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044683540216382766,
      "loss": 4.7633,
      "mean_token_accuracy": 0.22301261276006698,
      "num_tokens": 50640472.0,
      "step": 27520
    },
    {
      "entropy": 5.02932505607605,
      "epoch": 2.2991146007350483,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004468161824707883,
      "loss": 4.6702,
      "mean_token_accuracy": 0.23024013936519622,
      "num_tokens": 50649513.0,
      "step": 27525
    },
    {
      "entropy": 4.986678123474121,
      "epoch": 2.2995322418977615,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00044679695976989985,
      "loss": 4.7581,
      "mean_token_accuracy": 0.2219816818833351,
      "num_tokens": 50659028.0,
      "step": 27530
    },
    {
      "entropy": 5.092750024795532,
      "epoch": 2.2999498830604743,
      "grad_norm": 1.125,
      "learning_rate": 0.00044677773406149885,
      "loss": 4.7451,
      "mean_token_accuracy": 0.21795104146003724,
      "num_tokens": 50668888.0,
      "step": 27535
    },
    {
      "entropy": 5.032612276077271,
      "epoch": 2.3003675242231876,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044675850534592185,
      "loss": 4.7336,
      "mean_token_accuracy": 0.2259683519601822,
      "num_tokens": 50678094.0,
      "step": 27540
    },
    {
      "entropy": 4.932534551620483,
      "epoch": 2.3007851653859004,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044673927362350545,
      "loss": 4.6268,
      "mean_token_accuracy": 0.24029613882303238,
      "num_tokens": 50686692.0,
      "step": 27545
    },
    {
      "entropy": 5.1167387008667,
      "epoch": 2.3012028065486136,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044672003889458634,
      "loss": 4.8275,
      "mean_token_accuracy": 0.21605969220399857,
      "num_tokens": 50695579.0,
      "step": 27550
    },
    {
      "entropy": 5.023748302459717,
      "epoch": 2.3016204477113265,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044670080115950117,
      "loss": 4.7284,
      "mean_token_accuracy": 0.22127141207456588,
      "num_tokens": 50704205.0,
      "step": 27555
    },
    {
      "entropy": 5.099490594863892,
      "epoch": 2.3020380888740393,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044668156041858696,
      "loss": 4.778,
      "mean_token_accuracy": 0.21996843218803405,
      "num_tokens": 50713926.0,
      "step": 27560
    },
    {
      "entropy": 5.143434619903564,
      "epoch": 2.3024557300367525,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044666231667218025,
      "loss": 4.8038,
      "mean_token_accuracy": 0.2180517554283142,
      "num_tokens": 50723732.0,
      "step": 27565
    },
    {
      "entropy": 4.990884447097779,
      "epoch": 2.3028733711994653,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044664306992061816,
      "loss": 4.6554,
      "mean_token_accuracy": 0.2254351034760475,
      "num_tokens": 50732572.0,
      "step": 27570
    },
    {
      "entropy": 5.045852184295654,
      "epoch": 2.3032910123621786,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044662382016423747,
      "loss": 4.8105,
      "mean_token_accuracy": 0.2182328313589096,
      "num_tokens": 50741145.0,
      "step": 27575
    },
    {
      "entropy": 5.147225093841553,
      "epoch": 2.3037086535248914,
      "grad_norm": 1.125,
      "learning_rate": 0.0004466045674033752,
      "loss": 4.8001,
      "mean_token_accuracy": 0.219646418094635,
      "num_tokens": 50749966.0,
      "step": 27580
    },
    {
      "entropy": 5.062183570861817,
      "epoch": 2.304126294687604,
      "grad_norm": 1.125,
      "learning_rate": 0.0004465853116383684,
      "loss": 4.7608,
      "mean_token_accuracy": 0.22207331657409668,
      "num_tokens": 50759263.0,
      "step": 27585
    },
    {
      "entropy": 4.987922477722168,
      "epoch": 2.3045439358503175,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044656605286955424,
      "loss": 4.6871,
      "mean_token_accuracy": 0.2236202225089073,
      "num_tokens": 50768071.0,
      "step": 27590
    },
    {
      "entropy": 5.055246877670288,
      "epoch": 2.3049615770130303,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004465467910972698,
      "loss": 4.7293,
      "mean_token_accuracy": 0.227569642663002,
      "num_tokens": 50777658.0,
      "step": 27595
    },
    {
      "entropy": 5.120436811447144,
      "epoch": 2.3053792181757435,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004465275263218523,
      "loss": 4.82,
      "mean_token_accuracy": 0.21221221834421158,
      "num_tokens": 50788438.0,
      "step": 27600
    },
    {
      "entropy": 5.0065278053283695,
      "epoch": 2.3057968593384564,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044650825854363895,
      "loss": 4.7336,
      "mean_token_accuracy": 0.22694420218467712,
      "num_tokens": 50797459.0,
      "step": 27605
    },
    {
      "entropy": 4.996972227096558,
      "epoch": 2.306214500501169,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004464889877629671,
      "loss": 4.6291,
      "mean_token_accuracy": 0.2315104678273201,
      "num_tokens": 50806088.0,
      "step": 27610
    },
    {
      "entropy": 5.024400997161865,
      "epoch": 2.3066321416638824,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004464697139801741,
      "loss": 4.6919,
      "mean_token_accuracy": 0.2192596286535263,
      "num_tokens": 50815567.0,
      "step": 27615
    },
    {
      "entropy": 4.992121315002441,
      "epoch": 2.3070497828265952,
      "grad_norm": 1.03125,
      "learning_rate": 0.00044645043719559736,
      "loss": 4.7026,
      "mean_token_accuracy": 0.23344603627920152,
      "num_tokens": 50824519.0,
      "step": 27620
    },
    {
      "entropy": 5.076477193832398,
      "epoch": 2.3074674239893085,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044643115740957437,
      "loss": 4.8212,
      "mean_token_accuracy": 0.22226884216070175,
      "num_tokens": 50834875.0,
      "step": 27625
    },
    {
      "entropy": 5.075249862670899,
      "epoch": 2.3078850651520213,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004464118746224426,
      "loss": 4.7371,
      "mean_token_accuracy": 0.22677480727434157,
      "num_tokens": 50843668.0,
      "step": 27630
    },
    {
      "entropy": 5.15842113494873,
      "epoch": 2.3083027063147346,
      "grad_norm": 1.25,
      "learning_rate": 0.0004463925888345396,
      "loss": 4.846,
      "mean_token_accuracy": 0.2086987763643265,
      "num_tokens": 50853291.0,
      "step": 27635
    },
    {
      "entropy": 5.09701509475708,
      "epoch": 2.3087203474774474,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00044637330004620307,
      "loss": 4.8017,
      "mean_token_accuracy": 0.2145507514476776,
      "num_tokens": 50863392.0,
      "step": 27640
    },
    {
      "entropy": 5.091030931472778,
      "epoch": 2.3091379886401606,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044635400825777065,
      "loss": 4.7605,
      "mean_token_accuracy": 0.2233569175004959,
      "num_tokens": 50871857.0,
      "step": 27645
    },
    {
      "entropy": 5.030451536178589,
      "epoch": 2.3095556298028734,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044633471346957996,
      "loss": 4.7572,
      "mean_token_accuracy": 0.22535109519958496,
      "num_tokens": 50882171.0,
      "step": 27650
    },
    {
      "entropy": 4.928315925598144,
      "epoch": 2.3099732709655862,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044631541568196896,
      "loss": 4.5527,
      "mean_token_accuracy": 0.24084637612104415,
      "num_tokens": 50890925.0,
      "step": 27655
    },
    {
      "entropy": 5.0181151866912845,
      "epoch": 2.3103909121282995,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044629611489527537,
      "loss": 4.7535,
      "mean_token_accuracy": 0.22582820802927017,
      "num_tokens": 50900555.0,
      "step": 27660
    },
    {
      "entropy": 4.9805420398712155,
      "epoch": 2.3108085532910123,
      "grad_norm": 1.125,
      "learning_rate": 0.0004462768111098371,
      "loss": 4.6039,
      "mean_token_accuracy": 0.23757093995809556,
      "num_tokens": 50908763.0,
      "step": 27665
    },
    {
      "entropy": 5.037858152389527,
      "epoch": 2.3112261944537256,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044625750432599213,
      "loss": 4.6312,
      "mean_token_accuracy": 0.23104348927736282,
      "num_tokens": 50916906.0,
      "step": 27670
    },
    {
      "entropy": 5.061362934112549,
      "epoch": 2.3116438356164384,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044623819454407844,
      "loss": 4.8589,
      "mean_token_accuracy": 0.20788802057504654,
      "num_tokens": 50926338.0,
      "step": 27675
    },
    {
      "entropy": 5.049060297012329,
      "epoch": 2.312061476779151,
      "grad_norm": 1.1875,
      "learning_rate": 0.000446218881764434,
      "loss": 4.7655,
      "mean_token_accuracy": 0.2260572835803032,
      "num_tokens": 50935392.0,
      "step": 27680
    },
    {
      "entropy": 5.100816440582276,
      "epoch": 2.3124791179418644,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044619956598739686,
      "loss": 4.7921,
      "mean_token_accuracy": 0.21659348756074906,
      "num_tokens": 50945078.0,
      "step": 27685
    },
    {
      "entropy": 5.0829010009765625,
      "epoch": 2.3128967591045773,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044618024721330523,
      "loss": 4.7502,
      "mean_token_accuracy": 0.2197419673204422,
      "num_tokens": 50955091.0,
      "step": 27690
    },
    {
      "entropy": 5.066631317138672,
      "epoch": 2.3133144002672905,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004461609254424974,
      "loss": 4.8003,
      "mean_token_accuracy": 0.2228017583489418,
      "num_tokens": 50964730.0,
      "step": 27695
    },
    {
      "entropy": 5.018654632568359,
      "epoch": 2.3137320414300033,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044614160067531157,
      "loss": 4.7055,
      "mean_token_accuracy": 0.22572572082281112,
      "num_tokens": 50973666.0,
      "step": 27700
    },
    {
      "entropy": 5.102194786071777,
      "epoch": 2.314149682592716,
      "grad_norm": 1.1875,
      "learning_rate": 0.000446122272912086,
      "loss": 4.7935,
      "mean_token_accuracy": 0.2155224695801735,
      "num_tokens": 50983339.0,
      "step": 27705
    },
    {
      "entropy": 5.018773221969605,
      "epoch": 2.3145673237554294,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000446102942153159,
      "loss": 4.7394,
      "mean_token_accuracy": 0.22656601965427398,
      "num_tokens": 50993563.0,
      "step": 27710
    },
    {
      "entropy": 5.078161096572876,
      "epoch": 2.314984964918142,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004460836083988691,
      "loss": 4.7763,
      "mean_token_accuracy": 0.2186690554022789,
      "num_tokens": 51002179.0,
      "step": 27715
    },
    {
      "entropy": 5.018532466888428,
      "epoch": 2.3154026060808555,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044606427164955466,
      "loss": 4.6604,
      "mean_token_accuracy": 0.2323373556137085,
      "num_tokens": 51010770.0,
      "step": 27720
    },
    {
      "entropy": 4.95811095237732,
      "epoch": 2.3158202472435683,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004460449319055543,
      "loss": 4.7344,
      "mean_token_accuracy": 0.23065850436687468,
      "num_tokens": 51021224.0,
      "step": 27725
    },
    {
      "entropy": 5.049547624588013,
      "epoch": 2.316237888406281,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004460255891672064,
      "loss": 4.6764,
      "mean_token_accuracy": 0.22530232071876527,
      "num_tokens": 51030425.0,
      "step": 27730
    },
    {
      "entropy": 5.122429990768433,
      "epoch": 2.3166555295689943,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044600624343484985,
      "loss": 4.8442,
      "mean_token_accuracy": 0.21956423074007034,
      "num_tokens": 51040135.0,
      "step": 27735
    },
    {
      "entropy": 5.068213605880738,
      "epoch": 2.317073170731707,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004459868947088231,
      "loss": 4.8803,
      "mean_token_accuracy": 0.20836528539657592,
      "num_tokens": 51050696.0,
      "step": 27740
    },
    {
      "entropy": 5.096569299697876,
      "epoch": 2.3174908118944204,
      "grad_norm": 1.015625,
      "learning_rate": 0.00044596754298946494,
      "loss": 4.7248,
      "mean_token_accuracy": 0.2214997738599777,
      "num_tokens": 51060092.0,
      "step": 27745
    },
    {
      "entropy": 5.068529367446899,
      "epoch": 2.3179084530571332,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044594818827711416,
      "loss": 4.6768,
      "mean_token_accuracy": 0.22531329989433288,
      "num_tokens": 51069139.0,
      "step": 27750
    },
    {
      "entropy": 4.923473167419433,
      "epoch": 2.3183260942198465,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004459288305721095,
      "loss": 4.6484,
      "mean_token_accuracy": 0.22811872661113738,
      "num_tokens": 51078262.0,
      "step": 27755
    },
    {
      "entropy": 5.020704746246338,
      "epoch": 2.3187437353825593,
      "grad_norm": 1.015625,
      "learning_rate": 0.00044590946987479,
      "loss": 4.7614,
      "mean_token_accuracy": 0.22222097218036652,
      "num_tokens": 51088679.0,
      "step": 27760
    },
    {
      "entropy": 5.015578174591065,
      "epoch": 2.3191613765452725,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004458901061854944,
      "loss": 4.6627,
      "mean_token_accuracy": 0.23361112028360367,
      "num_tokens": 51097673.0,
      "step": 27765
    },
    {
      "entropy": 5.052057409286499,
      "epoch": 2.3195790177079854,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044587073950456193,
      "loss": 4.7179,
      "mean_token_accuracy": 0.2266784280538559,
      "num_tokens": 51107546.0,
      "step": 27770
    },
    {
      "entropy": 5.060459852218628,
      "epoch": 2.319996658870698,
      "grad_norm": 1.125,
      "learning_rate": 0.0004458513698323314,
      "loss": 4.7883,
      "mean_token_accuracy": 0.22057192027568817,
      "num_tokens": 51116860.0,
      "step": 27775
    },
    {
      "entropy": 5.055401134490967,
      "epoch": 2.3204143000334114,
      "grad_norm": 1.140625,
      "learning_rate": 0.000445831997169142,
      "loss": 4.7594,
      "mean_token_accuracy": 0.23068131059408187,
      "num_tokens": 51125815.0,
      "step": 27780
    },
    {
      "entropy": 5.06870927810669,
      "epoch": 2.3208319411961242,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004458126215153329,
      "loss": 4.7537,
      "mean_token_accuracy": 0.219576632976532,
      "num_tokens": 51134561.0,
      "step": 27785
    },
    {
      "entropy": 5.049367666244507,
      "epoch": 2.3212495823588375,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044579324287124324,
      "loss": 4.7369,
      "mean_token_accuracy": 0.21862749010324478,
      "num_tokens": 51143506.0,
      "step": 27790
    },
    {
      "entropy": 5.141071510314942,
      "epoch": 2.3216672235215503,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044577386123721225,
      "loss": 4.7575,
      "mean_token_accuracy": 0.22267535030841829,
      "num_tokens": 51152632.0,
      "step": 27795
    },
    {
      "entropy": 5.104702186584473,
      "epoch": 2.322084864684263,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044575447661357925,
      "loss": 4.8608,
      "mean_token_accuracy": 0.2126636654138565,
      "num_tokens": 51162947.0,
      "step": 27800
    },
    {
      "entropy": 5.0334327697753904,
      "epoch": 2.3225025058469764,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044573508900068367,
      "loss": 4.7413,
      "mean_token_accuracy": 0.22588728964328766,
      "num_tokens": 51172030.0,
      "step": 27805
    },
    {
      "entropy": 4.978287553787231,
      "epoch": 2.322920147009689,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004457156983988648,
      "loss": 4.6821,
      "mean_token_accuracy": 0.22398641705513,
      "num_tokens": 51181009.0,
      "step": 27810
    },
    {
      "entropy": 5.0823822021484375,
      "epoch": 2.3233377881724024,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044569630480846214,
      "loss": 4.6474,
      "mean_token_accuracy": 0.2326285645365715,
      "num_tokens": 51189731.0,
      "step": 27815
    },
    {
      "entropy": 5.037945127487182,
      "epoch": 2.3237554293351153,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004456769082298152,
      "loss": 4.7288,
      "mean_token_accuracy": 0.2216989740729332,
      "num_tokens": 51199726.0,
      "step": 27820
    },
    {
      "entropy": 4.970191574096679,
      "epoch": 2.324173070497828,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044565750866326355,
      "loss": 4.6875,
      "mean_token_accuracy": 0.2297812357544899,
      "num_tokens": 51209068.0,
      "step": 27825
    },
    {
      "entropy": 5.02950496673584,
      "epoch": 2.3245907116605413,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004456381061091467,
      "loss": 4.6644,
      "mean_token_accuracy": 0.2291696012020111,
      "num_tokens": 51217766.0,
      "step": 27830
    },
    {
      "entropy": 5.036939191818237,
      "epoch": 2.325008352823254,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044561870056780445,
      "loss": 4.7689,
      "mean_token_accuracy": 0.22008754909038544,
      "num_tokens": 51227134.0,
      "step": 27835
    },
    {
      "entropy": 5.021519947052002,
      "epoch": 2.3254259939859674,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004455992920395766,
      "loss": 4.7233,
      "mean_token_accuracy": 0.2261600226163864,
      "num_tokens": 51236059.0,
      "step": 27840
    },
    {
      "entropy": 5.162261629104615,
      "epoch": 2.32584363514868,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044557988052480263,
      "loss": 4.866,
      "mean_token_accuracy": 0.21271420866250992,
      "num_tokens": 51245730.0,
      "step": 27845
    },
    {
      "entropy": 5.094819974899292,
      "epoch": 2.326261276311393,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044556046602382257,
      "loss": 4.6767,
      "mean_token_accuracy": 0.2226028710603714,
      "num_tokens": 51254633.0,
      "step": 27850
    },
    {
      "entropy": 5.029579591751099,
      "epoch": 2.3266789174741063,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004455410485369763,
      "loss": 4.7985,
      "mean_token_accuracy": 0.21637978851795198,
      "num_tokens": 51264264.0,
      "step": 27855
    },
    {
      "entropy": 5.013226747512817,
      "epoch": 2.327096558636819,
      "grad_norm": 1.125,
      "learning_rate": 0.0004455216280646037,
      "loss": 4.7884,
      "mean_token_accuracy": 0.21276647895574569,
      "num_tokens": 51273376.0,
      "step": 27860
    },
    {
      "entropy": 4.965868139266968,
      "epoch": 2.3275141997995323,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004455022046070446,
      "loss": 4.6774,
      "mean_token_accuracy": 0.2256431519985199,
      "num_tokens": 51282623.0,
      "step": 27865
    },
    {
      "entropy": 5.006729602813721,
      "epoch": 2.327931840962245,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044548277816463925,
      "loss": 4.7151,
      "mean_token_accuracy": 0.22337645739316941,
      "num_tokens": 51291543.0,
      "step": 27870
    },
    {
      "entropy": 5.0606142520904545,
      "epoch": 2.3283494821249584,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044546334873772777,
      "loss": 4.7261,
      "mean_token_accuracy": 0.22038601636886596,
      "num_tokens": 51300455.0,
      "step": 27875
    },
    {
      "entropy": 4.96059284210205,
      "epoch": 2.328767123287671,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044544391632665004,
      "loss": 4.5672,
      "mean_token_accuracy": 0.2284917101264,
      "num_tokens": 51308871.0,
      "step": 27880
    },
    {
      "entropy": 4.988801908493042,
      "epoch": 2.3291847644503845,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004454244809317465,
      "loss": 4.619,
      "mean_token_accuracy": 0.22119584828615188,
      "num_tokens": 51318132.0,
      "step": 27885
    },
    {
      "entropy": 5.097054672241211,
      "epoch": 2.3296024056130973,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044540504255335717,
      "loss": 4.7122,
      "mean_token_accuracy": 0.22185055315494537,
      "num_tokens": 51326573.0,
      "step": 27890
    },
    {
      "entropy": 5.025275468826294,
      "epoch": 2.33002004677581,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004453856011918225,
      "loss": 4.723,
      "mean_token_accuracy": 0.22175590246915816,
      "num_tokens": 51336139.0,
      "step": 27895
    },
    {
      "entropy": 5.016311740875244,
      "epoch": 2.3304376879385234,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044536615684748286,
      "loss": 4.744,
      "mean_token_accuracy": 0.2240385890007019,
      "num_tokens": 51345163.0,
      "step": 27900
    },
    {
      "entropy": 5.0260584354400635,
      "epoch": 2.330855329101236,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004453467095206785,
      "loss": 4.7358,
      "mean_token_accuracy": 0.2279016777873039,
      "num_tokens": 51354305.0,
      "step": 27905
    },
    {
      "entropy": 5.001116609573364,
      "epoch": 2.3312729702639494,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004453272592117499,
      "loss": 4.627,
      "mean_token_accuracy": 0.23353002071380616,
      "num_tokens": 51362732.0,
      "step": 27910
    },
    {
      "entropy": 4.981220722198486,
      "epoch": 2.3316906114266622,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004453078059210377,
      "loss": 4.7483,
      "mean_token_accuracy": 0.2268858954310417,
      "num_tokens": 51372802.0,
      "step": 27915
    },
    {
      "entropy": 4.993206596374511,
      "epoch": 2.332108252589375,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044528834964888234,
      "loss": 4.6438,
      "mean_token_accuracy": 0.22530897110700607,
      "num_tokens": 51382618.0,
      "step": 27920
    },
    {
      "entropy": 5.030665254592895,
      "epoch": 2.3325258937520883,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044526889039562444,
      "loss": 4.7247,
      "mean_token_accuracy": 0.22868942022323607,
      "num_tokens": 51392089.0,
      "step": 27925
    },
    {
      "entropy": 5.087162733078003,
      "epoch": 2.332943534914801,
      "grad_norm": 1.125,
      "learning_rate": 0.00044524942816160463,
      "loss": 4.7055,
      "mean_token_accuracy": 0.2225024878978729,
      "num_tokens": 51401323.0,
      "step": 27930
    },
    {
      "entropy": 5.046551084518432,
      "epoch": 2.3333611760775144,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044522996294716374,
      "loss": 4.7454,
      "mean_token_accuracy": 0.22721018344163896,
      "num_tokens": 51410079.0,
      "step": 27935
    },
    {
      "entropy": 4.962905073165894,
      "epoch": 2.333778817240227,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044521049475264237,
      "loss": 4.6946,
      "mean_token_accuracy": 0.21976813077926635,
      "num_tokens": 51418638.0,
      "step": 27940
    },
    {
      "entropy": 5.084222078323364,
      "epoch": 2.33419645840294,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004451910235783814,
      "loss": 4.9257,
      "mean_token_accuracy": 0.2073317736387253,
      "num_tokens": 51428766.0,
      "step": 27945
    },
    {
      "entropy": 5.103308200836182,
      "epoch": 2.3346140995656532,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044517154942472166,
      "loss": 4.8055,
      "mean_token_accuracy": 0.22076202034950257,
      "num_tokens": 51437828.0,
      "step": 27950
    },
    {
      "entropy": 5.092401838302612,
      "epoch": 2.335031740728366,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004451520722920042,
      "loss": 4.7758,
      "mean_token_accuracy": 0.22273098975419997,
      "num_tokens": 51448070.0,
      "step": 27955
    },
    {
      "entropy": 5.0391065120697025,
      "epoch": 2.3354493818910793,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004451325921805698,
      "loss": 4.6617,
      "mean_token_accuracy": 0.2255108192563057,
      "num_tokens": 51457990.0,
      "step": 27960
    },
    {
      "entropy": 5.036547517776489,
      "epoch": 2.335867023053792,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004451131090907597,
      "loss": 4.7134,
      "mean_token_accuracy": 0.22516053915023804,
      "num_tokens": 51466323.0,
      "step": 27965
    },
    {
      "entropy": 5.057635068893433,
      "epoch": 2.336284664216505,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004450936230229148,
      "loss": 4.753,
      "mean_token_accuracy": 0.21725132167339326,
      "num_tokens": 51476958.0,
      "step": 27970
    },
    {
      "entropy": 4.9902900695800785,
      "epoch": 2.336702305379218,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004450741339773763,
      "loss": 4.7423,
      "mean_token_accuracy": 0.22637540251016616,
      "num_tokens": 51486843.0,
      "step": 27975
    },
    {
      "entropy": 5.047124862670898,
      "epoch": 2.337119946541931,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044505464195448535,
      "loss": 4.6812,
      "mean_token_accuracy": 0.22782655358314513,
      "num_tokens": 51496331.0,
      "step": 27980
    },
    {
      "entropy": 5.07754955291748,
      "epoch": 2.3375375877046443,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044503514695458325,
      "loss": 4.7933,
      "mean_token_accuracy": 0.22377043068408967,
      "num_tokens": 51505693.0,
      "step": 27985
    },
    {
      "entropy": 5.070644664764404,
      "epoch": 2.337955228867357,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004450156489780112,
      "loss": 4.7156,
      "mean_token_accuracy": 0.2288290575146675,
      "num_tokens": 51514753.0,
      "step": 27990
    },
    {
      "entropy": 5.084746313095093,
      "epoch": 2.3383728700300703,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004449961480251106,
      "loss": 4.7603,
      "mean_token_accuracy": 0.21763782799243928,
      "num_tokens": 51524082.0,
      "step": 27995
    },
    {
      "entropy": 4.952900362014771,
      "epoch": 2.338790511192783,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004449766440962227,
      "loss": 4.6966,
      "mean_token_accuracy": 0.2245427116751671,
      "num_tokens": 51533437.0,
      "step": 28000
    },
    {
      "entropy": 4.991678476333618,
      "epoch": 2.3392081523554964,
      "grad_norm": 1.125,
      "learning_rate": 0.0004449571371916891,
      "loss": 4.6286,
      "mean_token_accuracy": 0.23021902441978453,
      "num_tokens": 51542612.0,
      "step": 28005
    },
    {
      "entropy": 5.0166727066040036,
      "epoch": 2.339625793518209,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004449376273118513,
      "loss": 4.6546,
      "mean_token_accuracy": 0.23098323345184327,
      "num_tokens": 51551292.0,
      "step": 28010
    },
    {
      "entropy": 5.083548974990845,
      "epoch": 2.340043434680922,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004449181144570507,
      "loss": 4.7643,
      "mean_token_accuracy": 0.22629761546850205,
      "num_tokens": 51560203.0,
      "step": 28015
    },
    {
      "entropy": 5.075332164764404,
      "epoch": 2.3404610758436353,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000444898598627629,
      "loss": 4.8011,
      "mean_token_accuracy": 0.21984476447105408,
      "num_tokens": 51569716.0,
      "step": 28020
    },
    {
      "entropy": 5.047201490402221,
      "epoch": 2.340878717006348,
      "grad_norm": 1.25,
      "learning_rate": 0.0004448790798239278,
      "loss": 4.7976,
      "mean_token_accuracy": 0.21388614773750306,
      "num_tokens": 51578409.0,
      "step": 28025
    },
    {
      "entropy": 5.111255502700805,
      "epoch": 2.3412963581690613,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044485955804628884,
      "loss": 4.8477,
      "mean_token_accuracy": 0.22658416479825974,
      "num_tokens": 51587846.0,
      "step": 28030
    },
    {
      "entropy": 5.06061019897461,
      "epoch": 2.341713999331774,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044484003329505384,
      "loss": 4.7541,
      "mean_token_accuracy": 0.22202427685260773,
      "num_tokens": 51596393.0,
      "step": 28035
    },
    {
      "entropy": 5.123487424850464,
      "epoch": 2.342131640494487,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004448205055705647,
      "loss": 4.7469,
      "mean_token_accuracy": 0.2251642495393753,
      "num_tokens": 51605521.0,
      "step": 28040
    },
    {
      "entropy": 5.07806715965271,
      "epoch": 2.3425492816572,
      "grad_norm": 1.125,
      "learning_rate": 0.0004448009748731631,
      "loss": 4.7599,
      "mean_token_accuracy": 0.21937748342752456,
      "num_tokens": 51614770.0,
      "step": 28045
    },
    {
      "entropy": 5.037444543838501,
      "epoch": 2.342966922819913,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004447814412031911,
      "loss": 4.7993,
      "mean_token_accuracy": 0.22665003985166549,
      "num_tokens": 51623249.0,
      "step": 28050
    },
    {
      "entropy": 5.064686012268067,
      "epoch": 2.3433845639826263,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044476190456099057,
      "loss": 4.7481,
      "mean_token_accuracy": 0.22551929354667663,
      "num_tokens": 51632974.0,
      "step": 28055
    },
    {
      "entropy": 5.113229703903198,
      "epoch": 2.343802205145339,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004447423649469036,
      "loss": 4.7079,
      "mean_token_accuracy": 0.21935662180185317,
      "num_tokens": 51641204.0,
      "step": 28060
    },
    {
      "entropy": 4.912143230438232,
      "epoch": 2.344219846308052,
      "grad_norm": 1.125,
      "learning_rate": 0.0004447228223612722,
      "loss": 4.5836,
      "mean_token_accuracy": 0.23029733449220657,
      "num_tokens": 51650316.0,
      "step": 28065
    },
    {
      "entropy": 5.046591472625733,
      "epoch": 2.344637487470765,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004447032768044385,
      "loss": 4.7756,
      "mean_token_accuracy": 0.22488242983818055,
      "num_tokens": 51659164.0,
      "step": 28070
    },
    {
      "entropy": 5.062238645553589,
      "epoch": 2.345055128633478,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004446837282767447,
      "loss": 4.7629,
      "mean_token_accuracy": 0.21751794666051866,
      "num_tokens": 51667982.0,
      "step": 28075
    },
    {
      "entropy": 5.093807172775269,
      "epoch": 2.3454727697961912,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044466417677853294,
      "loss": 4.7666,
      "mean_token_accuracy": 0.22434917688369752,
      "num_tokens": 51677397.0,
      "step": 28080
    },
    {
      "entropy": 5.068066740036011,
      "epoch": 2.345890410958904,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004446446223101456,
      "loss": 4.8726,
      "mean_token_accuracy": 0.21558971107006072,
      "num_tokens": 51686462.0,
      "step": 28085
    },
    {
      "entropy": 5.135827493667603,
      "epoch": 2.346308052121617,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044462506487192487,
      "loss": 4.7813,
      "mean_token_accuracy": 0.2135440781712532,
      "num_tokens": 51696654.0,
      "step": 28090
    },
    {
      "entropy": 5.0945068359375,
      "epoch": 2.34672569328433,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044460550446421323,
      "loss": 4.8251,
      "mean_token_accuracy": 0.21483394652605056,
      "num_tokens": 51706005.0,
      "step": 28095
    },
    {
      "entropy": 5.025141668319702,
      "epoch": 2.347143334447043,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000444585941087353,
      "loss": 4.7658,
      "mean_token_accuracy": 0.22008075714111328,
      "num_tokens": 51715543.0,
      "step": 28100
    },
    {
      "entropy": 5.012063932418823,
      "epoch": 2.347560975609756,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004445663747416868,
      "loss": 4.7381,
      "mean_token_accuracy": 0.22269109189510344,
      "num_tokens": 51723599.0,
      "step": 28105
    },
    {
      "entropy": 5.135872793197632,
      "epoch": 2.347978616772469,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044454680542755715,
      "loss": 4.8401,
      "mean_token_accuracy": 0.21722056567668915,
      "num_tokens": 51732850.0,
      "step": 28110
    },
    {
      "entropy": 5.118642330169678,
      "epoch": 2.3483962579351823,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004445272331453066,
      "loss": 4.7906,
      "mean_token_accuracy": 0.21954194754362105,
      "num_tokens": 51742601.0,
      "step": 28115
    },
    {
      "entropy": 5.084225177764893,
      "epoch": 2.348813899097895,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044450765789527777,
      "loss": 4.7754,
      "mean_token_accuracy": 0.21312471032142638,
      "num_tokens": 51751675.0,
      "step": 28120
    },
    {
      "entropy": 5.114529085159302,
      "epoch": 2.3492315402606083,
      "grad_norm": 1.0,
      "learning_rate": 0.0004444880796778133,
      "loss": 4.8679,
      "mean_token_accuracy": 0.21520100384950638,
      "num_tokens": 51761122.0,
      "step": 28125
    },
    {
      "entropy": 5.172735214233398,
      "epoch": 2.349649181423321,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000444468498493256,
      "loss": 4.7932,
      "mean_token_accuracy": 0.21959329694509505,
      "num_tokens": 51770337.0,
      "step": 28130
    },
    {
      "entropy": 5.010454893112183,
      "epoch": 2.350066822586034,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044444891434194863,
      "loss": 4.7639,
      "mean_token_accuracy": 0.21856583654880524,
      "num_tokens": 51779630.0,
      "step": 28135
    },
    {
      "entropy": 4.998806715011597,
      "epoch": 2.350484463748747,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004444293272242341,
      "loss": 4.6037,
      "mean_token_accuracy": 0.23949741423130036,
      "num_tokens": 51788885.0,
      "step": 28140
    },
    {
      "entropy": 5.0466100692749025,
      "epoch": 2.35090210491146,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044440973714045517,
      "loss": 4.7916,
      "mean_token_accuracy": 0.21037744879722595,
      "num_tokens": 51798152.0,
      "step": 28145
    },
    {
      "entropy": 5.05491189956665,
      "epoch": 2.3513197460741733,
      "grad_norm": 1.03125,
      "learning_rate": 0.00044439014409095495,
      "loss": 4.7542,
      "mean_token_accuracy": 0.22578264772891998,
      "num_tokens": 51807853.0,
      "step": 28150
    },
    {
      "entropy": 5.080895900726318,
      "epoch": 2.351737387236886,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004443705480760763,
      "loss": 4.8294,
      "mean_token_accuracy": 0.22208307832479476,
      "num_tokens": 51817403.0,
      "step": 28155
    },
    {
      "entropy": 5.098609256744385,
      "epoch": 2.352155028399599,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044435094909616243,
      "loss": 4.7356,
      "mean_token_accuracy": 0.21452984660863877,
      "num_tokens": 51826527.0,
      "step": 28160
    },
    {
      "entropy": 5.2011065006256105,
      "epoch": 2.352572669562312,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004443313471515562,
      "loss": 4.8798,
      "mean_token_accuracy": 0.2104501396417618,
      "num_tokens": 51835743.0,
      "step": 28165
    },
    {
      "entropy": 5.035329151153564,
      "epoch": 2.352990310725025,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044431174224260106,
      "loss": 4.7737,
      "mean_token_accuracy": 0.2143006294965744,
      "num_tokens": 51845465.0,
      "step": 28170
    },
    {
      "entropy": 5.031054449081421,
      "epoch": 2.353407951887738,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044429213436964,
      "loss": 4.7022,
      "mean_token_accuracy": 0.22274691313505174,
      "num_tokens": 51855242.0,
      "step": 28175
    },
    {
      "entropy": 5.047136926651001,
      "epoch": 2.353825593050451,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00044427252353301637,
      "loss": 4.6709,
      "mean_token_accuracy": 0.2386828675866127,
      "num_tokens": 51864142.0,
      "step": 28180
    },
    {
      "entropy": 5.0513725757598875,
      "epoch": 2.354243234213164,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044425290973307345,
      "loss": 4.6934,
      "mean_token_accuracy": 0.22563268542289733,
      "num_tokens": 51873555.0,
      "step": 28185
    },
    {
      "entropy": 4.936326265335083,
      "epoch": 2.354660875375877,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044423329297015463,
      "loss": 4.6661,
      "mean_token_accuracy": 0.2297448217868805,
      "num_tokens": 51882629.0,
      "step": 28190
    },
    {
      "entropy": 5.112296867370605,
      "epoch": 2.35507851653859,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044421367324460335,
      "loss": 4.7785,
      "mean_token_accuracy": 0.21724137514829636,
      "num_tokens": 51891639.0,
      "step": 28195
    },
    {
      "entropy": 5.027942180633545,
      "epoch": 2.355496157701303,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004441940505567629,
      "loss": 4.6159,
      "mean_token_accuracy": 0.2295992538332939,
      "num_tokens": 51900187.0,
      "step": 28200
    },
    {
      "entropy": 5.08958067893982,
      "epoch": 2.355913798864016,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044417442490697704,
      "loss": 4.7228,
      "mean_token_accuracy": 0.22207202315330504,
      "num_tokens": 51909002.0,
      "step": 28205
    },
    {
      "entropy": 5.10555739402771,
      "epoch": 2.356331440026729,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044415479629558924,
      "loss": 4.883,
      "mean_token_accuracy": 0.210508731007576,
      "num_tokens": 51919066.0,
      "step": 28210
    },
    {
      "entropy": 5.0880567073822025,
      "epoch": 2.356749081189442,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044413516472294315,
      "loss": 4.7828,
      "mean_token_accuracy": 0.21875430792570114,
      "num_tokens": 51928266.0,
      "step": 28215
    },
    {
      "entropy": 5.109641170501709,
      "epoch": 2.357166722352155,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004441155301893823,
      "loss": 4.8373,
      "mean_token_accuracy": 0.21623432338237764,
      "num_tokens": 51937646.0,
      "step": 28220
    },
    {
      "entropy": 5.037914896011353,
      "epoch": 2.357584363514868,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004440958926952506,
      "loss": 4.5679,
      "mean_token_accuracy": 0.23949397653341292,
      "num_tokens": 51945802.0,
      "step": 28225
    },
    {
      "entropy": 5.057547616958618,
      "epoch": 2.358002004677581,
      "grad_norm": 1.125,
      "learning_rate": 0.0004440762522408917,
      "loss": 4.8139,
      "mean_token_accuracy": 0.21482281535863876,
      "num_tokens": 51956024.0,
      "step": 28230
    },
    {
      "entropy": 5.056564283370972,
      "epoch": 2.358419645840294,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004440566088266496,
      "loss": 4.7291,
      "mean_token_accuracy": 0.2193291127681732,
      "num_tokens": 51965299.0,
      "step": 28235
    },
    {
      "entropy": 5.099196672439575,
      "epoch": 2.358837287003007,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00044403696245286803,
      "loss": 4.718,
      "mean_token_accuracy": 0.22559650540351867,
      "num_tokens": 51974138.0,
      "step": 28240
    },
    {
      "entropy": 5.005287170410156,
      "epoch": 2.3592549281657202,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044401731311989086,
      "loss": 4.7267,
      "mean_token_accuracy": 0.2260284200310707,
      "num_tokens": 51983770.0,
      "step": 28245
    },
    {
      "entropy": 5.0031884670257565,
      "epoch": 2.359672569328433,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004439976608280623,
      "loss": 4.7647,
      "mean_token_accuracy": 0.21695071160793306,
      "num_tokens": 51992294.0,
      "step": 28250
    },
    {
      "entropy": 5.198188924789429,
      "epoch": 2.360090210491146,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044397800557772614,
      "loss": 4.7705,
      "mean_token_accuracy": 0.21925611644983292,
      "num_tokens": 52001605.0,
      "step": 28255
    },
    {
      "entropy": 5.073822116851806,
      "epoch": 2.360507851653859,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044395834736922675,
      "loss": 4.6518,
      "mean_token_accuracy": 0.22964186370372772,
      "num_tokens": 52010335.0,
      "step": 28260
    },
    {
      "entropy": 4.978861856460571,
      "epoch": 2.360925492816572,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004439386862029081,
      "loss": 4.6412,
      "mean_token_accuracy": 0.23283127695322037,
      "num_tokens": 52020170.0,
      "step": 28265
    },
    {
      "entropy": 5.002046060562134,
      "epoch": 2.361343133979285,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004439190220791143,
      "loss": 4.7353,
      "mean_token_accuracy": 0.22323000580072402,
      "num_tokens": 52030917.0,
      "step": 28270
    },
    {
      "entropy": 5.024945020675659,
      "epoch": 2.361760775141998,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044389935499818974,
      "loss": 4.7754,
      "mean_token_accuracy": 0.2228654444217682,
      "num_tokens": 52041172.0,
      "step": 28275
    },
    {
      "entropy": 5.096514654159546,
      "epoch": 2.362178416304711,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004438796849604786,
      "loss": 4.7712,
      "mean_token_accuracy": 0.2222086012363434,
      "num_tokens": 52050955.0,
      "step": 28280
    },
    {
      "entropy": 5.052090120315552,
      "epoch": 2.362596057467424,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044386001196632536,
      "loss": 4.7109,
      "mean_token_accuracy": 0.22735912799835206,
      "num_tokens": 52060196.0,
      "step": 28285
    },
    {
      "entropy": 5.068576908111572,
      "epoch": 2.363013698630137,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004438403360160744,
      "loss": 4.7421,
      "mean_token_accuracy": 0.21858202815055847,
      "num_tokens": 52069375.0,
      "step": 28290
    },
    {
      "entropy": 5.069584751129151,
      "epoch": 2.36343133979285,
      "grad_norm": 1.265625,
      "learning_rate": 0.00044382065711007,
      "loss": 4.7059,
      "mean_token_accuracy": 0.2292953684926033,
      "num_tokens": 52078209.0,
      "step": 28295
    },
    {
      "entropy": 4.99881501197815,
      "epoch": 2.363848980955563,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044380097524865684,
      "loss": 4.7108,
      "mean_token_accuracy": 0.2296600565314293,
      "num_tokens": 52086822.0,
      "step": 28300
    },
    {
      "entropy": 5.084364032745361,
      "epoch": 2.3642666221182758,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044378129043217947,
      "loss": 4.7984,
      "mean_token_accuracy": 0.22277631610631943,
      "num_tokens": 52096011.0,
      "step": 28305
    },
    {
      "entropy": 5.066801643371582,
      "epoch": 2.364684263280989,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004437616026609824,
      "loss": 4.7148,
      "mean_token_accuracy": 0.22635398656129838,
      "num_tokens": 52104348.0,
      "step": 28310
    },
    {
      "entropy": 5.07302680015564,
      "epoch": 2.365101904443702,
      "grad_norm": 1.234375,
      "learning_rate": 0.00044374191193541037,
      "loss": 4.7188,
      "mean_token_accuracy": 0.2248107299208641,
      "num_tokens": 52113083.0,
      "step": 28315
    },
    {
      "entropy": 4.932910966873169,
      "epoch": 2.365519545606415,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044372221825580803,
      "loss": 4.5785,
      "mean_token_accuracy": 0.23560035973787308,
      "num_tokens": 52121726.0,
      "step": 28320
    },
    {
      "entropy": 5.033581399917603,
      "epoch": 2.365937186769128,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044370252162252016,
      "loss": 4.7759,
      "mean_token_accuracy": 0.2263714239001274,
      "num_tokens": 52131515.0,
      "step": 28325
    },
    {
      "entropy": 5.108557891845703,
      "epoch": 2.366354827931841,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004436828220358916,
      "loss": 4.7024,
      "mean_token_accuracy": 0.22424090802669525,
      "num_tokens": 52140252.0,
      "step": 28330
    },
    {
      "entropy": 5.051519680023193,
      "epoch": 2.366772469094554,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004436631194962672,
      "loss": 4.848,
      "mean_token_accuracy": 0.22405953705310822,
      "num_tokens": 52150353.0,
      "step": 28335
    },
    {
      "entropy": 4.967749786376953,
      "epoch": 2.3671901102572668,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044364341400399186,
      "loss": 4.6903,
      "mean_token_accuracy": 0.22222989350557326,
      "num_tokens": 52160280.0,
      "step": 28340
    },
    {
      "entropy": 5.165478754043579,
      "epoch": 2.36760775141998,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044362370555941057,
      "loss": 4.8669,
      "mean_token_accuracy": 0.20833866149187089,
      "num_tokens": 52169989.0,
      "step": 28345
    },
    {
      "entropy": 5.1133819103240965,
      "epoch": 2.368025392582693,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004436039941628684,
      "loss": 4.6588,
      "mean_token_accuracy": 0.22705061584711075,
      "num_tokens": 52179535.0,
      "step": 28350
    },
    {
      "entropy": 4.9182164669036865,
      "epoch": 2.368443033745406,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004435842798147103,
      "loss": 4.5749,
      "mean_token_accuracy": 0.24372731745243073,
      "num_tokens": 52188809.0,
      "step": 28355
    },
    {
      "entropy": 4.989161252975464,
      "epoch": 2.368860674908119,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044356456251528145,
      "loss": 4.6926,
      "mean_token_accuracy": 0.22629936933517455,
      "num_tokens": 52197758.0,
      "step": 28360
    },
    {
      "entropy": 4.979665660858155,
      "epoch": 2.369278316070832,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004435448422649271,
      "loss": 4.6469,
      "mean_token_accuracy": 0.22602479457855223,
      "num_tokens": 52206461.0,
      "step": 28365
    },
    {
      "entropy": 5.044564485549927,
      "epoch": 2.369695957233545,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004435251190639923,
      "loss": 4.7324,
      "mean_token_accuracy": 0.22129555344581603,
      "num_tokens": 52215786.0,
      "step": 28370
    },
    {
      "entropy": 5.094570970535278,
      "epoch": 2.370113598396258,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044350539291282247,
      "loss": 4.7317,
      "mean_token_accuracy": 0.2307875707745552,
      "num_tokens": 52224668.0,
      "step": 28375
    },
    {
      "entropy": 5.109680986404419,
      "epoch": 2.370531239558971,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004434856638117629,
      "loss": 4.7517,
      "mean_token_accuracy": 0.22036397606134414,
      "num_tokens": 52233999.0,
      "step": 28380
    },
    {
      "entropy": 4.955208826065063,
      "epoch": 2.370948880721684,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044346593176115904,
      "loss": 4.6551,
      "mean_token_accuracy": 0.23293425440788268,
      "num_tokens": 52243015.0,
      "step": 28385
    },
    {
      "entropy": 4.9366453170776365,
      "epoch": 2.371366521884397,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044344619676135624,
      "loss": 4.6033,
      "mean_token_accuracy": 0.2395635426044464,
      "num_tokens": 52251748.0,
      "step": 28390
    },
    {
      "entropy": 5.029508018493653,
      "epoch": 2.37178416304711,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004434264588127,
      "loss": 4.6484,
      "mean_token_accuracy": 0.22961161881685258,
      "num_tokens": 52261228.0,
      "step": 28395
    },
    {
      "entropy": 5.059555196762085,
      "epoch": 2.3722018042098227,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044340671791553586,
      "loss": 4.7717,
      "mean_token_accuracy": 0.22613440901041032,
      "num_tokens": 52270464.0,
      "step": 28400
    },
    {
      "entropy": 5.06751389503479,
      "epoch": 2.372619445372536,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004433869740702094,
      "loss": 4.7199,
      "mean_token_accuracy": 0.2220579355955124,
      "num_tokens": 52279888.0,
      "step": 28405
    },
    {
      "entropy": 5.003015518188477,
      "epoch": 2.373037086535249,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004433672272770663,
      "loss": 4.658,
      "mean_token_accuracy": 0.23575784265995026,
      "num_tokens": 52288618.0,
      "step": 28410
    },
    {
      "entropy": 5.013562393188477,
      "epoch": 2.373454727697962,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044334747753645224,
      "loss": 4.7831,
      "mean_token_accuracy": 0.21410464942455293,
      "num_tokens": 52298001.0,
      "step": 28415
    },
    {
      "entropy": 5.074120903015137,
      "epoch": 2.373872368860675,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004433277248487129,
      "loss": 4.7331,
      "mean_token_accuracy": 0.22474218308925628,
      "num_tokens": 52306863.0,
      "step": 28420
    },
    {
      "entropy": 5.129256868362427,
      "epoch": 2.3742900100233877,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004433079692141941,
      "loss": 4.6701,
      "mean_token_accuracy": 0.225442473590374,
      "num_tokens": 52315445.0,
      "step": 28425
    },
    {
      "entropy": 5.0016984939575195,
      "epoch": 2.374707651186101,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044328821063324177,
      "loss": 4.6979,
      "mean_token_accuracy": 0.22493379414081574,
      "num_tokens": 52324478.0,
      "step": 28430
    },
    {
      "entropy": 5.004116201400757,
      "epoch": 2.3751252923488138,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004432684491062017,
      "loss": 4.752,
      "mean_token_accuracy": 0.22436780482530594,
      "num_tokens": 52333136.0,
      "step": 28435
    },
    {
      "entropy": 5.103477430343628,
      "epoch": 2.375542933511527,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044324868463341987,
      "loss": 4.8526,
      "mean_token_accuracy": 0.21472582072019578,
      "num_tokens": 52343490.0,
      "step": 28440
    },
    {
      "entropy": 5.076018810272217,
      "epoch": 2.37596057467424,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044322891721524233,
      "loss": 4.7799,
      "mean_token_accuracy": 0.22279694974422454,
      "num_tokens": 52352628.0,
      "step": 28445
    },
    {
      "entropy": 5.055836534500122,
      "epoch": 2.376378215836953,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044320914685201516,
      "loss": 4.7001,
      "mean_token_accuracy": 0.22744858562946318,
      "num_tokens": 52361947.0,
      "step": 28450
    },
    {
      "entropy": 5.077479028701783,
      "epoch": 2.376795856999666,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004431893735440843,
      "loss": 4.7503,
      "mean_token_accuracy": 0.2169675648212433,
      "num_tokens": 52371251.0,
      "step": 28455
    },
    {
      "entropy": 5.036704111099243,
      "epoch": 2.3772134981623787,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044316959729179603,
      "loss": 4.6793,
      "mean_token_accuracy": 0.23629861176013947,
      "num_tokens": 52380681.0,
      "step": 28460
    },
    {
      "entropy": 5.043752956390381,
      "epoch": 2.377631139325092,
      "grad_norm": 1.234375,
      "learning_rate": 0.00044314981809549656,
      "loss": 4.6903,
      "mean_token_accuracy": 0.2240574389696121,
      "num_tokens": 52390059.0,
      "step": 28465
    },
    {
      "entropy": 5.062661790847779,
      "epoch": 2.3780487804878048,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004431300359555321,
      "loss": 4.7754,
      "mean_token_accuracy": 0.21863425374031067,
      "num_tokens": 52399574.0,
      "step": 28470
    },
    {
      "entropy": 5.110264873504638,
      "epoch": 2.378466421650518,
      "grad_norm": 1.171875,
      "learning_rate": 0.000443110250872249,
      "loss": 4.8586,
      "mean_token_accuracy": 0.21127362698316574,
      "num_tokens": 52408926.0,
      "step": 28475
    },
    {
      "entropy": 5.0882374286651615,
      "epoch": 2.378884062813231,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004430904628459935,
      "loss": 4.8048,
      "mean_token_accuracy": 0.2208354115486145,
      "num_tokens": 52418052.0,
      "step": 28480
    },
    {
      "entropy": 5.049007177352905,
      "epoch": 2.379301703975944,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044307067187711227,
      "loss": 4.687,
      "mean_token_accuracy": 0.22633873969316481,
      "num_tokens": 52427422.0,
      "step": 28485
    },
    {
      "entropy": 5.050131797790527,
      "epoch": 2.379719345138657,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004430508779659516,
      "loss": 4.7619,
      "mean_token_accuracy": 0.2231649339199066,
      "num_tokens": 52436535.0,
      "step": 28490
    },
    {
      "entropy": 5.037318086624145,
      "epoch": 2.3801369863013697,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044303108111285795,
      "loss": 4.7638,
      "mean_token_accuracy": 0.2232379734516144,
      "num_tokens": 52445640.0,
      "step": 28495
    },
    {
      "entropy": 5.097168970108032,
      "epoch": 2.380554627464083,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044301128131817805,
      "loss": 4.8411,
      "mean_token_accuracy": 0.2216578394174576,
      "num_tokens": 52454715.0,
      "step": 28500
    },
    {
      "entropy": 5.125636100769043,
      "epoch": 2.380972268626796,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044299147858225835,
      "loss": 4.836,
      "mean_token_accuracy": 0.21926853358745574,
      "num_tokens": 52464015.0,
      "step": 28505
    },
    {
      "entropy": 5.072052001953125,
      "epoch": 2.381389909789509,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004429716729054456,
      "loss": 4.7798,
      "mean_token_accuracy": 0.21937213689088822,
      "num_tokens": 52472927.0,
      "step": 28510
    },
    {
      "entropy": 5.1064377784729,
      "epoch": 2.381807550952222,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004429518642880866,
      "loss": 4.8432,
      "mean_token_accuracy": 0.21636544466018676,
      "num_tokens": 52482417.0,
      "step": 28515
    },
    {
      "entropy": 5.084608888626098,
      "epoch": 2.3822251921149347,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044293205273052805,
      "loss": 4.7868,
      "mean_token_accuracy": 0.2248980164527893,
      "num_tokens": 52491294.0,
      "step": 28520
    },
    {
      "entropy": 5.130961561203003,
      "epoch": 2.382642833277648,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004429122382331168,
      "loss": 4.7803,
      "mean_token_accuracy": 0.22296638786792755,
      "num_tokens": 52500253.0,
      "step": 28525
    },
    {
      "entropy": 5.024791240692139,
      "epoch": 2.3830604744403607,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044289242079619955,
      "loss": 4.7256,
      "mean_token_accuracy": 0.22827023416757583,
      "num_tokens": 52509828.0,
      "step": 28530
    },
    {
      "entropy": 5.015624904632569,
      "epoch": 2.383478115603074,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044287260042012353,
      "loss": 4.6359,
      "mean_token_accuracy": 0.2265006735920906,
      "num_tokens": 52518781.0,
      "step": 28535
    },
    {
      "entropy": 5.019326496124267,
      "epoch": 2.383895756765787,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004428527771052355,
      "loss": 4.7127,
      "mean_token_accuracy": 0.22036436647176744,
      "num_tokens": 52528102.0,
      "step": 28540
    },
    {
      "entropy": 5.141286134719849,
      "epoch": 2.3843133979284996,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044283295085188263,
      "loss": 4.7941,
      "mean_token_accuracy": 0.21874644309282304,
      "num_tokens": 52537609.0,
      "step": 28545
    },
    {
      "entropy": 5.0266156673431395,
      "epoch": 2.384731039091213,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044281312166041196,
      "loss": 4.7535,
      "mean_token_accuracy": 0.22419884353876113,
      "num_tokens": 52547169.0,
      "step": 28550
    },
    {
      "entropy": 4.9668288230896,
      "epoch": 2.3851486802539257,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004427932895311705,
      "loss": 4.7004,
      "mean_token_accuracy": 0.21974488645792006,
      "num_tokens": 52556220.0,
      "step": 28555
    },
    {
      "entropy": 5.049795770645142,
      "epoch": 2.385566321416639,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004427734544645055,
      "loss": 4.8507,
      "mean_token_accuracy": 0.2173231989145279,
      "num_tokens": 52565736.0,
      "step": 28560
    },
    {
      "entropy": 5.135066652297974,
      "epoch": 2.3859839625793517,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044275361646076427,
      "loss": 4.7322,
      "mean_token_accuracy": 0.22035346031188965,
      "num_tokens": 52575222.0,
      "step": 28565
    },
    {
      "entropy": 5.0328083515167235,
      "epoch": 2.386401603742065,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044273377552029415,
      "loss": 4.6645,
      "mean_token_accuracy": 0.22955933958292007,
      "num_tokens": 52584726.0,
      "step": 28570
    },
    {
      "entropy": 5.062915468215943,
      "epoch": 2.386819244904778,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004427139316434423,
      "loss": 4.8553,
      "mean_token_accuracy": 0.22282471507787704,
      "num_tokens": 52595080.0,
      "step": 28575
    },
    {
      "entropy": 5.081250286102295,
      "epoch": 2.3872368860674906,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004426940848305563,
      "loss": 4.8347,
      "mean_token_accuracy": 0.21203957349061966,
      "num_tokens": 52604692.0,
      "step": 28580
    },
    {
      "entropy": 5.034412002563476,
      "epoch": 2.387654527230204,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004426742350819834,
      "loss": 4.7068,
      "mean_token_accuracy": 0.22102681547403336,
      "num_tokens": 52613697.0,
      "step": 28585
    },
    {
      "entropy": 5.053024101257324,
      "epoch": 2.3880721683929167,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044265438239807117,
      "loss": 4.7446,
      "mean_token_accuracy": 0.22132233232259751,
      "num_tokens": 52622048.0,
      "step": 28590
    },
    {
      "entropy": 4.9737344741821286,
      "epoch": 2.38848980955563,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004426345267791672,
      "loss": 4.6472,
      "mean_token_accuracy": 0.23338089138269424,
      "num_tokens": 52631939.0,
      "step": 28595
    },
    {
      "entropy": 5.001908493041992,
      "epoch": 2.3889074507183428,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044261466822561907,
      "loss": 4.6829,
      "mean_token_accuracy": 0.23221762031316756,
      "num_tokens": 52641651.0,
      "step": 28600
    },
    {
      "entropy": 5.048834323883057,
      "epoch": 2.389325091881056,
      "grad_norm": 1.125,
      "learning_rate": 0.0004425948067377744,
      "loss": 4.718,
      "mean_token_accuracy": 0.2224217101931572,
      "num_tokens": 52650863.0,
      "step": 28605
    },
    {
      "entropy": 5.030978441238403,
      "epoch": 2.389742733043769,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044257494231598094,
      "loss": 4.755,
      "mean_token_accuracy": 0.22223663032054902,
      "num_tokens": 52659892.0,
      "step": 28610
    },
    {
      "entropy": 5.116883230209351,
      "epoch": 2.3901603742064816,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004425550749605863,
      "loss": 4.7285,
      "mean_token_accuracy": 0.2254884198307991,
      "num_tokens": 52669381.0,
      "step": 28615
    },
    {
      "entropy": 5.172024822235107,
      "epoch": 2.390578015369195,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044253520467193855,
      "loss": 4.7806,
      "mean_token_accuracy": 0.22350225895643233,
      "num_tokens": 52678742.0,
      "step": 28620
    },
    {
      "entropy": 5.090850305557251,
      "epoch": 2.3909956565319077,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044251533145038525,
      "loss": 4.8094,
      "mean_token_accuracy": 0.2229001611471176,
      "num_tokens": 52688494.0,
      "step": 28625
    },
    {
      "entropy": 5.007509183883667,
      "epoch": 2.391413297694621,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004424954552962744,
      "loss": 4.6397,
      "mean_token_accuracy": 0.22965849488973616,
      "num_tokens": 52696778.0,
      "step": 28630
    },
    {
      "entropy": 4.99131965637207,
      "epoch": 2.3918309388573338,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044247557620995416,
      "loss": 4.7198,
      "mean_token_accuracy": 0.2272978752851486,
      "num_tokens": 52706031.0,
      "step": 28635
    },
    {
      "entropy": 5.03007698059082,
      "epoch": 2.3922485800200466,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004424556941917723,
      "loss": 4.7246,
      "mean_token_accuracy": 0.230196276307106,
      "num_tokens": 52715513.0,
      "step": 28640
    },
    {
      "entropy": 5.133295726776123,
      "epoch": 2.39266622118276,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044243580924207686,
      "loss": 4.8616,
      "mean_token_accuracy": 0.2111040785908699,
      "num_tokens": 52724622.0,
      "step": 28645
    },
    {
      "entropy": 5.003381872177124,
      "epoch": 2.3930838623454727,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004424159213612161,
      "loss": 4.7001,
      "mean_token_accuracy": 0.22628536820411682,
      "num_tokens": 52733879.0,
      "step": 28650
    },
    {
      "entropy": 4.938460206985473,
      "epoch": 2.393501503508186,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004423960305495381,
      "loss": 4.5671,
      "mean_token_accuracy": 0.24046126306056975,
      "num_tokens": 52742656.0,
      "step": 28655
    },
    {
      "entropy": 5.04323034286499,
      "epoch": 2.3939191446708987,
      "grad_norm": 1.0,
      "learning_rate": 0.0004423761368073911,
      "loss": 4.8252,
      "mean_token_accuracy": 0.21257415115833284,
      "num_tokens": 52753073.0,
      "step": 28660
    },
    {
      "entropy": 5.066107892990113,
      "epoch": 2.3943367858336115,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004423562401351234,
      "loss": 4.7803,
      "mean_token_accuracy": 0.2252670034766197,
      "num_tokens": 52761891.0,
      "step": 28665
    },
    {
      "entropy": 5.0230247497558596,
      "epoch": 2.394754426996325,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004423363405330833,
      "loss": 4.6881,
      "mean_token_accuracy": 0.2232733279466629,
      "num_tokens": 52772175.0,
      "step": 28670
    },
    {
      "entropy": 5.011983108520508,
      "epoch": 2.3951720681590376,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044231643800161897,
      "loss": 4.7908,
      "mean_token_accuracy": 0.21905535012483596,
      "num_tokens": 52783073.0,
      "step": 28675
    },
    {
      "entropy": 5.051026153564453,
      "epoch": 2.395589709321751,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004422965325410791,
      "loss": 4.6463,
      "mean_token_accuracy": 0.22404792457818984,
      "num_tokens": 52791949.0,
      "step": 28680
    },
    {
      "entropy": 4.935914325714111,
      "epoch": 2.3960073504844637,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044227662415181207,
      "loss": 4.6141,
      "mean_token_accuracy": 0.23056407570838927,
      "num_tokens": 52800912.0,
      "step": 28685
    },
    {
      "entropy": 5.035098123550415,
      "epoch": 2.396424991647177,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044225671283416634,
      "loss": 4.7381,
      "mean_token_accuracy": 0.22152356505393983,
      "num_tokens": 52810419.0,
      "step": 28690
    },
    {
      "entropy": 5.065309000015259,
      "epoch": 2.3968426328098897,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044223679858849057,
      "loss": 4.6824,
      "mean_token_accuracy": 0.22630833387374877,
      "num_tokens": 52819148.0,
      "step": 28695
    },
    {
      "entropy": 5.0512697219848635,
      "epoch": 2.3972602739726026,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004422168814151333,
      "loss": 4.8058,
      "mean_token_accuracy": 0.21675920337438584,
      "num_tokens": 52828494.0,
      "step": 28700
    },
    {
      "entropy": 4.988232803344727,
      "epoch": 2.397677915135316,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044219696131444326,
      "loss": 4.761,
      "mean_token_accuracy": 0.21884320080280303,
      "num_tokens": 52837279.0,
      "step": 28705
    },
    {
      "entropy": 5.096050262451172,
      "epoch": 2.3980955562980286,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044217703828676913,
      "loss": 4.7395,
      "mean_token_accuracy": 0.23009616285562515,
      "num_tokens": 52846319.0,
      "step": 28710
    },
    {
      "entropy": 5.030514812469482,
      "epoch": 2.398513197460742,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044215711233245973,
      "loss": 4.695,
      "mean_token_accuracy": 0.22601534724235534,
      "num_tokens": 52855005.0,
      "step": 28715
    },
    {
      "entropy": 5.055240869522095,
      "epoch": 2.3989308386234547,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044213718345186394,
      "loss": 4.8254,
      "mean_token_accuracy": 0.22121654450893402,
      "num_tokens": 52864805.0,
      "step": 28720
    },
    {
      "entropy": 5.0851246356964115,
      "epoch": 2.399348479786168,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00044211725164533045,
      "loss": 4.7431,
      "mean_token_accuracy": 0.23150790482759476,
      "num_tokens": 52873808.0,
      "step": 28725
    },
    {
      "entropy": 5.072906970977783,
      "epoch": 2.3997661209488808,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044209731691320837,
      "loss": 4.7308,
      "mean_token_accuracy": 0.22298979610204697,
      "num_tokens": 52882564.0,
      "step": 28730
    },
    {
      "entropy": 5.085605144500732,
      "epoch": 2.4001837621115936,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004420773792558467,
      "loss": 4.7972,
      "mean_token_accuracy": 0.21362327933311462,
      "num_tokens": 52891271.0,
      "step": 28735
    },
    {
      "entropy": 5.001752758026123,
      "epoch": 2.400601403274307,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004420574386735943,
      "loss": 4.6857,
      "mean_token_accuracy": 0.22176723629236222,
      "num_tokens": 52900075.0,
      "step": 28740
    },
    {
      "entropy": 4.981393575668335,
      "epoch": 2.4010190444370196,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044203749516680035,
      "loss": 4.6863,
      "mean_token_accuracy": 0.22566887736320496,
      "num_tokens": 52908744.0,
      "step": 28745
    },
    {
      "entropy": 4.999891042709351,
      "epoch": 2.401436685599733,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044201754873581396,
      "loss": 4.7166,
      "mean_token_accuracy": 0.23060406595468522,
      "num_tokens": 52918242.0,
      "step": 28750
    },
    {
      "entropy": 5.104976081848145,
      "epoch": 2.4018543267624457,
      "grad_norm": 1.25,
      "learning_rate": 0.00044199759938098437,
      "loss": 4.7951,
      "mean_token_accuracy": 0.22164206057786942,
      "num_tokens": 52927453.0,
      "step": 28755
    },
    {
      "entropy": 5.009840250015259,
      "epoch": 2.4022719679251585,
      "grad_norm": 1.125,
      "learning_rate": 0.00044197764710266087,
      "loss": 4.6216,
      "mean_token_accuracy": 0.2383708104491234,
      "num_tokens": 52935467.0,
      "step": 28760
    },
    {
      "entropy": 5.159610557556152,
      "epoch": 2.4026896090878718,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004419576919011926,
      "loss": 4.8325,
      "mean_token_accuracy": 0.22435345947742463,
      "num_tokens": 52944770.0,
      "step": 28765
    },
    {
      "entropy": 5.068985557556152,
      "epoch": 2.4031072502505846,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044193773377692886,
      "loss": 4.6846,
      "mean_token_accuracy": 0.22681887894868852,
      "num_tokens": 52953679.0,
      "step": 28770
    },
    {
      "entropy": 4.974546909332275,
      "epoch": 2.403524891413298,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044191777273021933,
      "loss": 4.675,
      "mean_token_accuracy": 0.22760800868272782,
      "num_tokens": 52962101.0,
      "step": 28775
    },
    {
      "entropy": 4.947806739807129,
      "epoch": 2.4039425325760106,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004418978087614131,
      "loss": 4.726,
      "mean_token_accuracy": 0.223204043507576,
      "num_tokens": 52971661.0,
      "step": 28780
    },
    {
      "entropy": 5.142502212524414,
      "epoch": 2.4043601737387235,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00044187784187085994,
      "loss": 4.7719,
      "mean_token_accuracy": 0.2222331017255783,
      "num_tokens": 52981091.0,
      "step": 28785
    },
    {
      "entropy": 5.145110082626343,
      "epoch": 2.4047778149014367,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004418578720589093,
      "loss": 4.806,
      "mean_token_accuracy": 0.2149713695049286,
      "num_tokens": 52990392.0,
      "step": 28790
    },
    {
      "entropy": 5.0755987644195555,
      "epoch": 2.4051954560641495,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004418378993259107,
      "loss": 4.7716,
      "mean_token_accuracy": 0.2200204536318779,
      "num_tokens": 52999203.0,
      "step": 28795
    },
    {
      "entropy": 5.194940423965454,
      "epoch": 2.405613097226863,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004418179236722139,
      "loss": 4.8682,
      "mean_token_accuracy": 0.22355751991271972,
      "num_tokens": 53009421.0,
      "step": 28800
    },
    {
      "entropy": 5.008869314193726,
      "epoch": 2.4060307383895756,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004417979450981685,
      "loss": 4.69,
      "mean_token_accuracy": 0.22269248366355895,
      "num_tokens": 53018968.0,
      "step": 28805
    },
    {
      "entropy": 5.106674480438232,
      "epoch": 2.406448379552289,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044177796360412436,
      "loss": 4.7516,
      "mean_token_accuracy": 0.22181342244148256,
      "num_tokens": 53027708.0,
      "step": 28810
    },
    {
      "entropy": 5.095004510879517,
      "epoch": 2.4068660207150017,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004417579791904312,
      "loss": 4.8082,
      "mean_token_accuracy": 0.22493516951799392,
      "num_tokens": 53036373.0,
      "step": 28815
    },
    {
      "entropy": 5.148033189773559,
      "epoch": 2.4072836618777145,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004417379918574388,
      "loss": 4.8532,
      "mean_token_accuracy": 0.22215977311134338,
      "num_tokens": 53046138.0,
      "step": 28820
    },
    {
      "entropy": 5.058585786819458,
      "epoch": 2.4077013030404277,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004417180016054972,
      "loss": 4.7291,
      "mean_token_accuracy": 0.21783749908208846,
      "num_tokens": 53055039.0,
      "step": 28825
    },
    {
      "entropy": 5.074195528030396,
      "epoch": 2.4081189442031405,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044169800843495637,
      "loss": 4.7396,
      "mean_token_accuracy": 0.22284829914569854,
      "num_tokens": 53064901.0,
      "step": 28830
    },
    {
      "entropy": 5.017162704467774,
      "epoch": 2.408536585365854,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004416780123461662,
      "loss": 4.7097,
      "mean_token_accuracy": 0.22283827811479567,
      "num_tokens": 53074138.0,
      "step": 28835
    },
    {
      "entropy": 5.020942115783692,
      "epoch": 2.4089542265285666,
      "grad_norm": 1.25,
      "learning_rate": 0.00044165801333947676,
      "loss": 4.6766,
      "mean_token_accuracy": 0.22248571515083312,
      "num_tokens": 53083020.0,
      "step": 28840
    },
    {
      "entropy": 5.005581426620483,
      "epoch": 2.40937186769128,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004416380114152382,
      "loss": 4.7071,
      "mean_token_accuracy": 0.2206792488694191,
      "num_tokens": 53092141.0,
      "step": 28845
    },
    {
      "entropy": 4.9962366104125975,
      "epoch": 2.4097895088539927,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044161800657380076,
      "loss": 4.7071,
      "mean_token_accuracy": 0.22541548907756806,
      "num_tokens": 53101305.0,
      "step": 28850
    },
    {
      "entropy": 5.071627855300903,
      "epoch": 2.4102071500167055,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004415979988155145,
      "loss": 4.7452,
      "mean_token_accuracy": 0.22360073477029802,
      "num_tokens": 53109946.0,
      "step": 28855
    },
    {
      "entropy": 5.111125707626343,
      "epoch": 2.4106247911794187,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004415779881407297,
      "loss": 4.7226,
      "mean_token_accuracy": 0.2255729615688324,
      "num_tokens": 53118369.0,
      "step": 28860
    },
    {
      "entropy": 5.060685825347901,
      "epoch": 2.4110424323421316,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004415579745497967,
      "loss": 4.7438,
      "mean_token_accuracy": 0.2218719244003296,
      "num_tokens": 53127389.0,
      "step": 28865
    },
    {
      "entropy": 5.058525323867798,
      "epoch": 2.411460073504845,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004415379580430659,
      "loss": 4.7654,
      "mean_token_accuracy": 0.2174159437417984,
      "num_tokens": 53136055.0,
      "step": 28870
    },
    {
      "entropy": 5.037100982666016,
      "epoch": 2.4118777146675576,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044151793862088766,
      "loss": 4.7346,
      "mean_token_accuracy": 0.22619975805282594,
      "num_tokens": 53145687.0,
      "step": 28875
    },
    {
      "entropy": 5.053214693069458,
      "epoch": 2.4122953558302704,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004414979162836125,
      "loss": 4.7666,
      "mean_token_accuracy": 0.21909843236207963,
      "num_tokens": 53155667.0,
      "step": 28880
    },
    {
      "entropy": 5.0007726669311525,
      "epoch": 2.4127129969929837,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004414778910315909,
      "loss": 4.6691,
      "mean_token_accuracy": 0.23145498782396318,
      "num_tokens": 53164615.0,
      "step": 28885
    },
    {
      "entropy": 5.0819355010986325,
      "epoch": 2.4131306381556965,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044145786286517343,
      "loss": 4.8031,
      "mean_token_accuracy": 0.21502113342285156,
      "num_tokens": 53174139.0,
      "step": 28890
    },
    {
      "entropy": 5.185388088226318,
      "epoch": 2.4135482793184098,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004414378317847107,
      "loss": 4.8004,
      "mean_token_accuracy": 0.21776794046163558,
      "num_tokens": 53182019.0,
      "step": 28895
    },
    {
      "entropy": 4.946096181869507,
      "epoch": 2.4139659204811226,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044141779779055334,
      "loss": 4.7048,
      "mean_token_accuracy": 0.2241524040699005,
      "num_tokens": 53191312.0,
      "step": 28900
    },
    {
      "entropy": 5.048893308639526,
      "epoch": 2.4143835616438354,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044139776088305215,
      "loss": 4.7109,
      "mean_token_accuracy": 0.2238335430622101,
      "num_tokens": 53200027.0,
      "step": 28905
    },
    {
      "entropy": 5.093730068206787,
      "epoch": 2.4148012028065486,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044137772106255793,
      "loss": 4.7542,
      "mean_token_accuracy": 0.2150525465607643,
      "num_tokens": 53209841.0,
      "step": 28910
    },
    {
      "entropy": 5.19576096534729,
      "epoch": 2.4152188439692615,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044135767832942136,
      "loss": 4.9354,
      "mean_token_accuracy": 0.21001821756362915,
      "num_tokens": 53219717.0,
      "step": 28915
    },
    {
      "entropy": 5.036483335494995,
      "epoch": 2.4156364851319747,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004413376326839934,
      "loss": 4.6527,
      "mean_token_accuracy": 0.22426954209804534,
      "num_tokens": 53228886.0,
      "step": 28920
    },
    {
      "entropy": 5.0899434089660645,
      "epoch": 2.4160541262946875,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044131758412662493,
      "loss": 4.7287,
      "mean_token_accuracy": 0.2253539264202118,
      "num_tokens": 53237314.0,
      "step": 28925
    },
    {
      "entropy": 4.988383960723877,
      "epoch": 2.4164717674574008,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000441297532657667,
      "loss": 4.6999,
      "mean_token_accuracy": 0.22711419612169265,
      "num_tokens": 53246502.0,
      "step": 28930
    },
    {
      "entropy": 5.01512508392334,
      "epoch": 2.4168894086201136,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004412774782774705,
      "loss": 4.734,
      "mean_token_accuracy": 0.21729044765233993,
      "num_tokens": 53256004.0,
      "step": 28935
    },
    {
      "entropy": 5.065680408477784,
      "epoch": 2.417307049782827,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004412574209863867,
      "loss": 4.7235,
      "mean_token_accuracy": 0.22541319727897643,
      "num_tokens": 53264301.0,
      "step": 28940
    },
    {
      "entropy": 5.0230193614959715,
      "epoch": 2.4177246909455397,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004412373607847666,
      "loss": 4.676,
      "mean_token_accuracy": 0.2312241017818451,
      "num_tokens": 53272664.0,
      "step": 28945
    },
    {
      "entropy": 5.075992155075073,
      "epoch": 2.4181423321082525,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044121729767296135,
      "loss": 4.7451,
      "mean_token_accuracy": 0.2215247705578804,
      "num_tokens": 53282210.0,
      "step": 28950
    },
    {
      "entropy": 5.044845247268677,
      "epoch": 2.4185599732709657,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044119723165132225,
      "loss": 4.8169,
      "mean_token_accuracy": 0.216656693816185,
      "num_tokens": 53291328.0,
      "step": 28955
    },
    {
      "entropy": 5.0551365375518795,
      "epoch": 2.4189776144336785,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044117716272020055,
      "loss": 4.749,
      "mean_token_accuracy": 0.2215956062078476,
      "num_tokens": 53301332.0,
      "step": 28960
    },
    {
      "entropy": 5.157235479354858,
      "epoch": 2.419395255596392,
      "grad_norm": 1.125,
      "learning_rate": 0.0004411570908799476,
      "loss": 4.7033,
      "mean_token_accuracy": 0.221449376642704,
      "num_tokens": 53310699.0,
      "step": 28965
    },
    {
      "entropy": 5.0235974311828615,
      "epoch": 2.4198128967591046,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044113701613091483,
      "loss": 4.7558,
      "mean_token_accuracy": 0.22242388427257537,
      "num_tokens": 53319760.0,
      "step": 28970
    },
    {
      "entropy": 4.99208550453186,
      "epoch": 2.4202305379218174,
      "grad_norm": 1.015625,
      "learning_rate": 0.00044111693847345354,
      "loss": 4.6476,
      "mean_token_accuracy": 0.22769930064678193,
      "num_tokens": 53328781.0,
      "step": 28975
    },
    {
      "entropy": 5.089646673202514,
      "epoch": 2.4206481790845307,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004410968579079153,
      "loss": 4.6703,
      "mean_token_accuracy": 0.23562841415405272,
      "num_tokens": 53337071.0,
      "step": 28980
    },
    {
      "entropy": 5.011759090423584,
      "epoch": 2.4210658202472435,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044107677443465165,
      "loss": 4.6873,
      "mean_token_accuracy": 0.2285304397344589,
      "num_tokens": 53346480.0,
      "step": 28985
    },
    {
      "entropy": 5.024772882461548,
      "epoch": 2.4214834614099567,
      "grad_norm": 1.0,
      "learning_rate": 0.00044105668805401415,
      "loss": 4.7659,
      "mean_token_accuracy": 0.2309776321053505,
      "num_tokens": 53355784.0,
      "step": 28990
    },
    {
      "entropy": 5.057003164291382,
      "epoch": 2.4219011025726696,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044103659876635443,
      "loss": 4.7469,
      "mean_token_accuracy": 0.22183727324008942,
      "num_tokens": 53364793.0,
      "step": 28995
    },
    {
      "entropy": 5.145279407501221,
      "epoch": 2.4223187437353824,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004410165065720242,
      "loss": 4.889,
      "mean_token_accuracy": 0.21675328761339188,
      "num_tokens": 53374473.0,
      "step": 29000
    },
    {
      "entropy": 5.043425989151001,
      "epoch": 2.4227363848980956,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004409964114713752,
      "loss": 4.7053,
      "mean_token_accuracy": 0.22809711545705796,
      "num_tokens": 53383453.0,
      "step": 29005
    },
    {
      "entropy": 5.115355110168457,
      "epoch": 2.4231540260608084,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004409763134647592,
      "loss": 4.7723,
      "mean_token_accuracy": 0.2210959881544113,
      "num_tokens": 53392628.0,
      "step": 29010
    },
    {
      "entropy": 5.078817415237427,
      "epoch": 2.4235716672235217,
      "grad_norm": 1.171875,
      "learning_rate": 0.000440956212552528,
      "loss": 4.7236,
      "mean_token_accuracy": 0.22534766644239426,
      "num_tokens": 53402614.0,
      "step": 29015
    },
    {
      "entropy": 5.139445018768311,
      "epoch": 2.4239893083862345,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004409361087350336,
      "loss": 4.9082,
      "mean_token_accuracy": 0.209831503033638,
      "num_tokens": 53412851.0,
      "step": 29020
    },
    {
      "entropy": 5.030457925796509,
      "epoch": 2.4244069495489473,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004409160020126279,
      "loss": 4.7008,
      "mean_token_accuracy": 0.2244071900844574,
      "num_tokens": 53422116.0,
      "step": 29025
    },
    {
      "entropy": 5.01342511177063,
      "epoch": 2.4248245907116606,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004408958923856628,
      "loss": 4.6531,
      "mean_token_accuracy": 0.23014004528522491,
      "num_tokens": 53431213.0,
      "step": 29030
    },
    {
      "entropy": 5.032641935348511,
      "epoch": 2.4252422318743734,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044087577985449047,
      "loss": 4.7181,
      "mean_token_accuracy": 0.23153737485408782,
      "num_tokens": 53439292.0,
      "step": 29035
    },
    {
      "entropy": 5.0190647602081295,
      "epoch": 2.4256598730370866,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004408556644194629,
      "loss": 4.721,
      "mean_token_accuracy": 0.22586775720119476,
      "num_tokens": 53447932.0,
      "step": 29040
    },
    {
      "entropy": 5.031076240539551,
      "epoch": 2.4260775141997994,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004408355460809323,
      "loss": 4.6506,
      "mean_token_accuracy": 0.22953959852457045,
      "num_tokens": 53457847.0,
      "step": 29045
    },
    {
      "entropy": 5.018021774291992,
      "epoch": 2.4264951553625127,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044081542483925086,
      "loss": 4.6416,
      "mean_token_accuracy": 0.22710852175951005,
      "num_tokens": 53466106.0,
      "step": 29050
    },
    {
      "entropy": 5.003336238861084,
      "epoch": 2.4269127965252255,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004407953006947708,
      "loss": 4.6682,
      "mean_token_accuracy": 0.23417604416608812,
      "num_tokens": 53474516.0,
      "step": 29055
    },
    {
      "entropy": 4.952232217788696,
      "epoch": 2.4273304376879388,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044077517364784455,
      "loss": 4.653,
      "mean_token_accuracy": 0.2276951014995575,
      "num_tokens": 53483640.0,
      "step": 29060
    },
    {
      "entropy": 5.05161600112915,
      "epoch": 2.4277480788506516,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044075504369882416,
      "loss": 4.7688,
      "mean_token_accuracy": 0.2266720488667488,
      "num_tokens": 53492987.0,
      "step": 29065
    },
    {
      "entropy": 5.100113391876221,
      "epoch": 2.4281657200133644,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004407349108480623,
      "loss": 4.7315,
      "mean_token_accuracy": 0.2199292853474617,
      "num_tokens": 53502319.0,
      "step": 29070
    },
    {
      "entropy": 5.076500511169433,
      "epoch": 2.4285833611760776,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044071477509591126,
      "loss": 4.7088,
      "mean_token_accuracy": 0.22352020293474198,
      "num_tokens": 53511739.0,
      "step": 29075
    },
    {
      "entropy": 5.100769424438477,
      "epoch": 2.4290010023387905,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004406946364427237,
      "loss": 4.7885,
      "mean_token_accuracy": 0.21963652074337006,
      "num_tokens": 53521369.0,
      "step": 29080
    },
    {
      "entropy": 5.054755926132202,
      "epoch": 2.4294186435015037,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044067449488885207,
      "loss": 4.6441,
      "mean_token_accuracy": 0.23809336125850677,
      "num_tokens": 53530697.0,
      "step": 29085
    },
    {
      "entropy": 4.987119913101196,
      "epoch": 2.4298362846642165,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044065435043464903,
      "loss": 4.725,
      "mean_token_accuracy": 0.22730336338281631,
      "num_tokens": 53539105.0,
      "step": 29090
    },
    {
      "entropy": 5.070553493499756,
      "epoch": 2.4302539258269293,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004406342030804671,
      "loss": 4.8565,
      "mean_token_accuracy": 0.21973638087511063,
      "num_tokens": 53548537.0,
      "step": 29095
    },
    {
      "entropy": 5.033302640914917,
      "epoch": 2.4306715669896426,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004406140528266591,
      "loss": 4.667,
      "mean_token_accuracy": 0.2290607824921608,
      "num_tokens": 53557467.0,
      "step": 29100
    },
    {
      "entropy": 5.107591819763184,
      "epoch": 2.4310892081523554,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044059389967357774,
      "loss": 4.8087,
      "mean_token_accuracy": 0.21915810108184813,
      "num_tokens": 53566970.0,
      "step": 29105
    },
    {
      "entropy": 5.045313358306885,
      "epoch": 2.4315068493150687,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044057374362157587,
      "loss": 4.7089,
      "mean_token_accuracy": 0.2214150458574295,
      "num_tokens": 53576320.0,
      "step": 29110
    },
    {
      "entropy": 4.99330587387085,
      "epoch": 2.4319244904777815,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044055358467100627,
      "loss": 4.6882,
      "mean_token_accuracy": 0.2311881273984909,
      "num_tokens": 53585135.0,
      "step": 29115
    },
    {
      "entropy": 5.016636848449707,
      "epoch": 2.4323421316404943,
      "grad_norm": 1.125,
      "learning_rate": 0.00044053342282222196,
      "loss": 4.7678,
      "mean_token_accuracy": 0.2264965817332268,
      "num_tokens": 53594138.0,
      "step": 29120
    },
    {
      "entropy": 5.011389636993409,
      "epoch": 2.4327597728032075,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004405132580755758,
      "loss": 4.6211,
      "mean_token_accuracy": 0.22938501685857773,
      "num_tokens": 53603133.0,
      "step": 29125
    },
    {
      "entropy": 5.053129768371582,
      "epoch": 2.4331774139659204,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044049309043142084,
      "loss": 4.7256,
      "mean_token_accuracy": 0.2218228191137314,
      "num_tokens": 53612219.0,
      "step": 29130
    },
    {
      "entropy": 5.043371248245239,
      "epoch": 2.4335950551286336,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004404729198901101,
      "loss": 4.7036,
      "mean_token_accuracy": 0.21433357298374175,
      "num_tokens": 53621548.0,
      "step": 29135
    },
    {
      "entropy": 5.16055850982666,
      "epoch": 2.4340126962913464,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044045274645199665,
      "loss": 4.839,
      "mean_token_accuracy": 0.21618059724569322,
      "num_tokens": 53631376.0,
      "step": 29140
    },
    {
      "entropy": 5.028030157089233,
      "epoch": 2.4344303374540592,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004404325701174338,
      "loss": 4.6578,
      "mean_token_accuracy": 0.23058862686157228,
      "num_tokens": 53639965.0,
      "step": 29145
    },
    {
      "entropy": 4.991183853149414,
      "epoch": 2.4348479786167725,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004404123908867747,
      "loss": 4.6017,
      "mean_token_accuracy": 0.2325405240058899,
      "num_tokens": 53649799.0,
      "step": 29150
    },
    {
      "entropy": 4.981632280349731,
      "epoch": 2.4352656197794853,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044039220876037255,
      "loss": 4.6932,
      "mean_token_accuracy": 0.23501036167144776,
      "num_tokens": 53658612.0,
      "step": 29155
    },
    {
      "entropy": 5.03657717704773,
      "epoch": 2.4356832609421986,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004403720237385808,
      "loss": 4.6805,
      "mean_token_accuracy": 0.2219746083021164,
      "num_tokens": 53667197.0,
      "step": 29160
    },
    {
      "entropy": 4.940001344680786,
      "epoch": 2.4361009021049114,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044035183582175264,
      "loss": 4.6094,
      "mean_token_accuracy": 0.23619025498628615,
      "num_tokens": 53675953.0,
      "step": 29165
    },
    {
      "entropy": 5.087270021438599,
      "epoch": 2.4365185432676246,
      "grad_norm": 1.125,
      "learning_rate": 0.00044033164501024153,
      "loss": 4.7986,
      "mean_token_accuracy": 0.22814014106988906,
      "num_tokens": 53685135.0,
      "step": 29170
    },
    {
      "entropy": 5.118917846679688,
      "epoch": 2.4369361844303374,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000440311451304401,
      "loss": 4.7622,
      "mean_token_accuracy": 0.21663580536842347,
      "num_tokens": 53694024.0,
      "step": 29175
    },
    {
      "entropy": 5.050327777862549,
      "epoch": 2.4373538255930507,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044029125470458445,
      "loss": 4.765,
      "mean_token_accuracy": 0.2191445603966713,
      "num_tokens": 53703699.0,
      "step": 29180
    },
    {
      "entropy": 4.947392368316651,
      "epoch": 2.4377714667557635,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004402710552111456,
      "loss": 4.6281,
      "mean_token_accuracy": 0.22763569056987762,
      "num_tokens": 53713121.0,
      "step": 29185
    },
    {
      "entropy": 4.984712028503418,
      "epoch": 2.4381891079184763,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044025085282443807,
      "loss": 4.6207,
      "mean_token_accuracy": 0.23171403408050537,
      "num_tokens": 53721896.0,
      "step": 29190
    },
    {
      "entropy": 5.0216490745544435,
      "epoch": 2.4386067490811896,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044023064754481535,
      "loss": 4.6643,
      "mean_token_accuracy": 0.22935260981321334,
      "num_tokens": 53731085.0,
      "step": 29195
    },
    {
      "entropy": 4.967660808563233,
      "epoch": 2.4390243902439024,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004402104393726313,
      "loss": 4.7026,
      "mean_token_accuracy": 0.23096876442432404,
      "num_tokens": 53739816.0,
      "step": 29200
    },
    {
      "entropy": 4.985525369644165,
      "epoch": 2.4394420314066156,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004401902283082396,
      "loss": 4.6802,
      "mean_token_accuracy": 0.22739672362804414,
      "num_tokens": 53748495.0,
      "step": 29205
    },
    {
      "entropy": 5.1562614917755125,
      "epoch": 2.4398596725693285,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044017001435199413,
      "loss": 4.8405,
      "mean_token_accuracy": 0.2188267007470131,
      "num_tokens": 53758622.0,
      "step": 29210
    },
    {
      "entropy": 5.088157701492309,
      "epoch": 2.4402773137320413,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004401497975042487,
      "loss": 4.8291,
      "mean_token_accuracy": 0.21140149980783463,
      "num_tokens": 53769867.0,
      "step": 29215
    },
    {
      "entropy": 5.1474662780761715,
      "epoch": 2.4406949548947545,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044012957776535735,
      "loss": 4.8934,
      "mean_token_accuracy": 0.21481760442256928,
      "num_tokens": 53780217.0,
      "step": 29220
    },
    {
      "entropy": 5.112566947937012,
      "epoch": 2.4411125960574673,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004401093551356739,
      "loss": 4.6814,
      "mean_token_accuracy": 0.2263025775551796,
      "num_tokens": 53788977.0,
      "step": 29225
    },
    {
      "entropy": 5.010116720199585,
      "epoch": 2.4415302372201806,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044008912961555246,
      "loss": 4.7591,
      "mean_token_accuracy": 0.22128846347332,
      "num_tokens": 53797955.0,
      "step": 29230
    },
    {
      "entropy": 4.964774179458618,
      "epoch": 2.4419478783828934,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004400689012053472,
      "loss": 4.7036,
      "mean_token_accuracy": 0.22987840473651885,
      "num_tokens": 53806929.0,
      "step": 29235
    },
    {
      "entropy": 5.0027000427246096,
      "epoch": 2.442365519545606,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000440048669905412,
      "loss": 4.7128,
      "mean_token_accuracy": 0.22546891868114471,
      "num_tokens": 53816071.0,
      "step": 29240
    },
    {
      "entropy": 5.077680063247681,
      "epoch": 2.4427831607083195,
      "grad_norm": 1.125,
      "learning_rate": 0.00044002843571610124,
      "loss": 4.6654,
      "mean_token_accuracy": 0.22587739825248718,
      "num_tokens": 53824376.0,
      "step": 29245
    },
    {
      "entropy": 4.993162345886231,
      "epoch": 2.4432008018710323,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000440008198637769,
      "loss": 4.6594,
      "mean_token_accuracy": 0.23390183001756668,
      "num_tokens": 53833504.0,
      "step": 29250
    },
    {
      "entropy": 4.99869966506958,
      "epoch": 2.4436184430337455,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004399879586707697,
      "loss": 4.7339,
      "mean_token_accuracy": 0.22542153745889665,
      "num_tokens": 53842027.0,
      "step": 29255
    },
    {
      "entropy": 5.040570402145386,
      "epoch": 2.4440360841964583,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004399677158154576,
      "loss": 4.7252,
      "mean_token_accuracy": 0.22675841003656388,
      "num_tokens": 53849895.0,
      "step": 29260
    },
    {
      "entropy": 5.077063083648682,
      "epoch": 2.444453725359171,
      "grad_norm": 1.015625,
      "learning_rate": 0.000439947470072187,
      "loss": 4.801,
      "mean_token_accuracy": 0.21473625749349595,
      "num_tokens": 53860030.0,
      "step": 29265
    },
    {
      "entropy": 5.080618095397949,
      "epoch": 2.4448713665218844,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004399272214413124,
      "loss": 4.7294,
      "mean_token_accuracy": 0.22476644068956375,
      "num_tokens": 53869298.0,
      "step": 29270
    },
    {
      "entropy": 5.029177618026734,
      "epoch": 2.4452890076845972,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004399069699231883,
      "loss": 4.68,
      "mean_token_accuracy": 0.22427801191806793,
      "num_tokens": 53878019.0,
      "step": 29275
    },
    {
      "entropy": 5.051843595504761,
      "epoch": 2.4457066488473105,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043988671551816914,
      "loss": 4.7243,
      "mean_token_accuracy": 0.22836442589759826,
      "num_tokens": 53886783.0,
      "step": 29280
    },
    {
      "entropy": 5.053199577331543,
      "epoch": 2.4461242900100233,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043986645822660956,
      "loss": 4.6952,
      "mean_token_accuracy": 0.23321400880813598,
      "num_tokens": 53895459.0,
      "step": 29285
    },
    {
      "entropy": 5.034732294082642,
      "epoch": 2.4465419311727366,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043984619804886423,
      "loss": 4.8043,
      "mean_token_accuracy": 0.22095730602741243,
      "num_tokens": 53904901.0,
      "step": 29290
    },
    {
      "entropy": 5.085344839096069,
      "epoch": 2.4469595723354494,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004398259349852877,
      "loss": 4.746,
      "mean_token_accuracy": 0.21937040835618973,
      "num_tokens": 53913170.0,
      "step": 29295
    },
    {
      "entropy": 4.994237279891967,
      "epoch": 2.4473772134981626,
      "grad_norm": 1.125,
      "learning_rate": 0.00043980566903623487,
      "loss": 4.5991,
      "mean_token_accuracy": 0.2320322424173355,
      "num_tokens": 53921950.0,
      "step": 29300
    },
    {
      "entropy": 5.056183576583862,
      "epoch": 2.4477948546608754,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004397854002020604,
      "loss": 4.7885,
      "mean_token_accuracy": 0.22352688610553742,
      "num_tokens": 53930984.0,
      "step": 29305
    },
    {
      "entropy": 5.110606241226196,
      "epoch": 2.4482124958235882,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043976512848311906,
      "loss": 4.7796,
      "mean_token_accuracy": 0.21978871822357177,
      "num_tokens": 53941498.0,
      "step": 29310
    },
    {
      "entropy": 4.994100141525268,
      "epoch": 2.4486301369863015,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004397448538797659,
      "loss": 4.6744,
      "mean_token_accuracy": 0.22991507798433303,
      "num_tokens": 53950229.0,
      "step": 29315
    },
    {
      "entropy": 5.081235742568969,
      "epoch": 2.4490477781490143,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004397245763923557,
      "loss": 4.803,
      "mean_token_accuracy": 0.2204243689775467,
      "num_tokens": 53959227.0,
      "step": 29320
    },
    {
      "entropy": 5.146136713027954,
      "epoch": 2.4494654193117276,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004397042960212436,
      "loss": 4.7795,
      "mean_token_accuracy": 0.21231137961149216,
      "num_tokens": 53967895.0,
      "step": 29325
    },
    {
      "entropy": 5.086907625198364,
      "epoch": 2.4498830604744404,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004396840127667845,
      "loss": 4.8346,
      "mean_token_accuracy": 0.21550584584474564,
      "num_tokens": 53977822.0,
      "step": 29330
    },
    {
      "entropy": 5.0721752643585205,
      "epoch": 2.450300701637153,
      "grad_norm": 1.125,
      "learning_rate": 0.0004396637266293335,
      "loss": 4.7078,
      "mean_token_accuracy": 0.22411493211984634,
      "num_tokens": 53986924.0,
      "step": 29335
    },
    {
      "entropy": 5.069673871994018,
      "epoch": 2.4507183427998664,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004396434376092458,
      "loss": 4.7083,
      "mean_token_accuracy": 0.22483818084001542,
      "num_tokens": 53996482.0,
      "step": 29340
    },
    {
      "entropy": 5.003226375579834,
      "epoch": 2.4511359839625793,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004396231457068765,
      "loss": 4.6424,
      "mean_token_accuracy": 0.23542011827230452,
      "num_tokens": 54005717.0,
      "step": 29345
    },
    {
      "entropy": 4.915382289886475,
      "epoch": 2.4515536251252925,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00043960285092258097,
      "loss": 4.5605,
      "mean_token_accuracy": 0.2411163255572319,
      "num_tokens": 54013825.0,
      "step": 29350
    },
    {
      "entropy": 5.02608094215393,
      "epoch": 2.4519712662880053,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004395825532567143,
      "loss": 4.8181,
      "mean_token_accuracy": 0.2204995334148407,
      "num_tokens": 54022975.0,
      "step": 29355
    },
    {
      "entropy": 5.095614814758301,
      "epoch": 2.452388907450718,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000439562252709632,
      "loss": 4.7008,
      "mean_token_accuracy": 0.2213215485215187,
      "num_tokens": 54031035.0,
      "step": 29360
    },
    {
      "entropy": 5.047044992446899,
      "epoch": 2.4528065486134314,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004395419492816894,
      "loss": 4.695,
      "mean_token_accuracy": 0.23056688159704208,
      "num_tokens": 54040424.0,
      "step": 29365
    },
    {
      "entropy": 4.993093252182007,
      "epoch": 2.453224189776144,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004395216429732418,
      "loss": 4.6836,
      "mean_token_accuracy": 0.22697857469320298,
      "num_tokens": 54049960.0,
      "step": 29370
    },
    {
      "entropy": 5.033384037017822,
      "epoch": 2.4536418309388575,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043950133378464485,
      "loss": 4.7303,
      "mean_token_accuracy": 0.2150715783238411,
      "num_tokens": 54058332.0,
      "step": 29375
    },
    {
      "entropy": 4.9659263610839846,
      "epoch": 2.4540594721015703,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004394810217162541,
      "loss": 4.5902,
      "mean_token_accuracy": 0.23611114025115967,
      "num_tokens": 54067049.0,
      "step": 29380
    },
    {
      "entropy": 4.996583700180054,
      "epoch": 2.454477113264283,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004394607067684251,
      "loss": 4.719,
      "mean_token_accuracy": 0.22787582576274873,
      "num_tokens": 54076200.0,
      "step": 29385
    },
    {
      "entropy": 5.081518363952637,
      "epoch": 2.4548947544269963,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043944038894151336,
      "loss": 4.6865,
      "mean_token_accuracy": 0.22673103958368301,
      "num_tokens": 54084374.0,
      "step": 29390
    },
    {
      "entropy": 5.052175140380859,
      "epoch": 2.455312395589709,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004394200682358748,
      "loss": 4.734,
      "mean_token_accuracy": 0.22548232227563858,
      "num_tokens": 54093618.0,
      "step": 29395
    },
    {
      "entropy": 5.058290243148804,
      "epoch": 2.4557300367524224,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004393997446518649,
      "loss": 4.7035,
      "mean_token_accuracy": 0.22868026793003082,
      "num_tokens": 54102942.0,
      "step": 29400
    },
    {
      "entropy": 4.990610122680664,
      "epoch": 2.456147677915135,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004393794181898397,
      "loss": 4.6222,
      "mean_token_accuracy": 0.23076875507831573,
      "num_tokens": 54111331.0,
      "step": 29405
    },
    {
      "entropy": 4.955358839035034,
      "epoch": 2.4565653190778485,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004393590888501548,
      "loss": 4.6441,
      "mean_token_accuracy": 0.2380052238702774,
      "num_tokens": 54119920.0,
      "step": 29410
    },
    {
      "entropy": 4.91044282913208,
      "epoch": 2.4569829602405613,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004393387566331663,
      "loss": 4.6224,
      "mean_token_accuracy": 0.23250891864299775,
      "num_tokens": 54129813.0,
      "step": 29415
    },
    {
      "entropy": 5.051840162277221,
      "epoch": 2.4574006014032745,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043931842153923,
      "loss": 4.6862,
      "mean_token_accuracy": 0.2296125888824463,
      "num_tokens": 54138369.0,
      "step": 29420
    },
    {
      "entropy": 5.049570941925049,
      "epoch": 2.4578182425659874,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000439298083568702,
      "loss": 4.6691,
      "mean_token_accuracy": 0.22610202580690383,
      "num_tokens": 54147622.0,
      "step": 29425
    },
    {
      "entropy": 5.043299150466919,
      "epoch": 2.4582358837287,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004392777427219382,
      "loss": 4.8119,
      "mean_token_accuracy": 0.22208011597394944,
      "num_tokens": 54156619.0,
      "step": 29430
    },
    {
      "entropy": 5.035516452789307,
      "epoch": 2.4586535248914134,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043925739899929484,
      "loss": 4.6468,
      "mean_token_accuracy": 0.2312850058078766,
      "num_tokens": 54166525.0,
      "step": 29435
    },
    {
      "entropy": 4.975691080093384,
      "epoch": 2.4590711660541262,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000439237052401128,
      "loss": 4.6737,
      "mean_token_accuracy": 0.2296462595462799,
      "num_tokens": 54174955.0,
      "step": 29440
    },
    {
      "entropy": 4.999472284317017,
      "epoch": 2.4594888072168395,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00043921670292779386,
      "loss": 4.7582,
      "mean_token_accuracy": 0.22248605340719224,
      "num_tokens": 54185389.0,
      "step": 29445
    },
    {
      "entropy": 5.029579067230225,
      "epoch": 2.4599064483795523,
      "grad_norm": 1.25,
      "learning_rate": 0.0004391963505796487,
      "loss": 4.6693,
      "mean_token_accuracy": 0.22962997406721114,
      "num_tokens": 54194797.0,
      "step": 29450
    },
    {
      "entropy": 5.065591859817505,
      "epoch": 2.460324089542265,
      "grad_norm": 1.125,
      "learning_rate": 0.0004391759953570487,
      "loss": 4.7537,
      "mean_token_accuracy": 0.22687097936868666,
      "num_tokens": 54203815.0,
      "step": 29455
    },
    {
      "entropy": 5.046030139923095,
      "epoch": 2.4607417307049784,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004391556372603503,
      "loss": 4.7708,
      "mean_token_accuracy": 0.21828621476888657,
      "num_tokens": 54213075.0,
      "step": 29460
    },
    {
      "entropy": 5.142192697525024,
      "epoch": 2.461159371867691,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043913527628990984,
      "loss": 4.8242,
      "mean_token_accuracy": 0.2157505989074707,
      "num_tokens": 54222551.0,
      "step": 29465
    },
    {
      "entropy": 5.019593572616577,
      "epoch": 2.4615770130304044,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043911491244608386,
      "loss": 4.6744,
      "mean_token_accuracy": 0.2276737630367279,
      "num_tokens": 54232294.0,
      "step": 29470
    },
    {
      "entropy": 5.012940454483032,
      "epoch": 2.4619946541931172,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004390945457292288,
      "loss": 4.7357,
      "mean_token_accuracy": 0.22487962692976,
      "num_tokens": 54241293.0,
      "step": 29475
    },
    {
      "entropy": 4.967526006698608,
      "epoch": 2.46241229535583,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004390741761397012,
      "loss": 4.6861,
      "mean_token_accuracy": 0.23037376403808593,
      "num_tokens": 54249801.0,
      "step": 29480
    },
    {
      "entropy": 5.015773916244507,
      "epoch": 2.4628299365185433,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043905380367785757,
      "loss": 4.6237,
      "mean_token_accuracy": 0.22905367463827134,
      "num_tokens": 54257747.0,
      "step": 29485
    },
    {
      "entropy": 5.014893388748169,
      "epoch": 2.463247577681256,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004390334283440547,
      "loss": 4.7348,
      "mean_token_accuracy": 0.2303760752081871,
      "num_tokens": 54266050.0,
      "step": 29490
    },
    {
      "entropy": 5.080104541778565,
      "epoch": 2.4636652188439694,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043901305013864923,
      "loss": 4.7847,
      "mean_token_accuracy": 0.22201624065637587,
      "num_tokens": 54275318.0,
      "step": 29495
    },
    {
      "entropy": 5.07238883972168,
      "epoch": 2.464082860006682,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043899266906199784,
      "loss": 4.7085,
      "mean_token_accuracy": 0.2373589277267456,
      "num_tokens": 54284623.0,
      "step": 29500
    },
    {
      "entropy": 5.090625619888305,
      "epoch": 2.464500501169395,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004389722851144574,
      "loss": 4.7913,
      "mean_token_accuracy": 0.2262405514717102,
      "num_tokens": 54293876.0,
      "step": 29505
    },
    {
      "entropy": 5.023708057403565,
      "epoch": 2.4649181423321083,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043895189829638485,
      "loss": 4.8193,
      "mean_token_accuracy": 0.2172710195183754,
      "num_tokens": 54303133.0,
      "step": 29510
    },
    {
      "entropy": 5.087375116348267,
      "epoch": 2.465335783494821,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004389315086081368,
      "loss": 4.7049,
      "mean_token_accuracy": 0.22368141412734985,
      "num_tokens": 54311725.0,
      "step": 29515
    },
    {
      "entropy": 5.032686948776245,
      "epoch": 2.4657534246575343,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043891111605007043,
      "loss": 4.6217,
      "mean_token_accuracy": 0.2404793009161949,
      "num_tokens": 54320268.0,
      "step": 29520
    },
    {
      "entropy": 4.94815411567688,
      "epoch": 2.466171065820247,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004388907206225427,
      "loss": 4.6663,
      "mean_token_accuracy": 0.23965757191181183,
      "num_tokens": 54328833.0,
      "step": 29525
    },
    {
      "entropy": 4.998134136199951,
      "epoch": 2.4665887069829604,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004388703223259106,
      "loss": 4.6613,
      "mean_token_accuracy": 0.23298247456550597,
      "num_tokens": 54337556.0,
      "step": 29530
    },
    {
      "entropy": 5.048830652236939,
      "epoch": 2.467006348145673,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043884992116053137,
      "loss": 4.7706,
      "mean_token_accuracy": 0.22321724146604538,
      "num_tokens": 54346170.0,
      "step": 29535
    },
    {
      "entropy": 4.989711618423462,
      "epoch": 2.4674239893083865,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043882951712676183,
      "loss": 4.6788,
      "mean_token_accuracy": 0.22829713076353073,
      "num_tokens": 54355328.0,
      "step": 29540
    },
    {
      "entropy": 4.975377464294434,
      "epoch": 2.4678416304710993,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043880911022495955,
      "loss": 4.6416,
      "mean_token_accuracy": 0.22629291117191314,
      "num_tokens": 54364445.0,
      "step": 29545
    },
    {
      "entropy": 5.041415643692017,
      "epoch": 2.468259271633812,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004387887004554816,
      "loss": 4.734,
      "mean_token_accuracy": 0.22456182837486266,
      "num_tokens": 54373407.0,
      "step": 29550
    },
    {
      "entropy": 4.928916311264038,
      "epoch": 2.4686769127965253,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004387682878186852,
      "loss": 4.5925,
      "mean_token_accuracy": 0.239089173078537,
      "num_tokens": 54381116.0,
      "step": 29555
    },
    {
      "entropy": 5.089337873458862,
      "epoch": 2.469094553959238,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004387478723149279,
      "loss": 4.8332,
      "mean_token_accuracy": 0.2157948687672615,
      "num_tokens": 54390174.0,
      "step": 29560
    },
    {
      "entropy": 5.0761199474334715,
      "epoch": 2.4695121951219514,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043872745394456697,
      "loss": 4.6861,
      "mean_token_accuracy": 0.22491036355495453,
      "num_tokens": 54399893.0,
      "step": 29565
    },
    {
      "entropy": 5.08342490196228,
      "epoch": 2.4699298362846642,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00043870703270795986,
      "loss": 4.8258,
      "mean_token_accuracy": 0.22180687487125397,
      "num_tokens": 54410037.0,
      "step": 29570
    },
    {
      "entropy": 5.041772842407227,
      "epoch": 2.470347477447377,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004386866086054641,
      "loss": 4.7004,
      "mean_token_accuracy": 0.22721717357635499,
      "num_tokens": 54419128.0,
      "step": 29575
    },
    {
      "entropy": 5.025700616836548,
      "epoch": 2.4707651186100903,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043866618163743723,
      "loss": 4.6233,
      "mean_token_accuracy": 0.23141295611858367,
      "num_tokens": 54428351.0,
      "step": 29580
    },
    {
      "entropy": 4.975736284255982,
      "epoch": 2.471182759772803,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043864575180423685,
      "loss": 4.6874,
      "mean_token_accuracy": 0.22729222923517228,
      "num_tokens": 54437728.0,
      "step": 29585
    },
    {
      "entropy": 5.107052230834961,
      "epoch": 2.4716004009355164,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043862531910622056,
      "loss": 4.8024,
      "mean_token_accuracy": 0.22057415246963502,
      "num_tokens": 54446504.0,
      "step": 29590
    },
    {
      "entropy": 5.1072217464447025,
      "epoch": 2.472018042098229,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043860488354374613,
      "loss": 4.772,
      "mean_token_accuracy": 0.22437286376953125,
      "num_tokens": 54456078.0,
      "step": 29595
    },
    {
      "entropy": 4.947205209732056,
      "epoch": 2.472435683260942,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043858444511717126,
      "loss": 4.6689,
      "mean_token_accuracy": 0.23340708613395691,
      "num_tokens": 54465713.0,
      "step": 29600
    },
    {
      "entropy": 4.9608536720275875,
      "epoch": 2.4728533244236552,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00043856400382685383,
      "loss": 4.6407,
      "mean_token_accuracy": 0.23152783513069153,
      "num_tokens": 54473803.0,
      "step": 29605
    },
    {
      "entropy": 5.078576326370239,
      "epoch": 2.473270965586368,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004385435596731516,
      "loss": 4.723,
      "mean_token_accuracy": 0.22739557772874833,
      "num_tokens": 54482920.0,
      "step": 29610
    },
    {
      "entropy": 5.053995990753174,
      "epoch": 2.4736886067490813,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043852311265642246,
      "loss": 4.6686,
      "mean_token_accuracy": 0.2270842134952545,
      "num_tokens": 54491556.0,
      "step": 29615
    },
    {
      "entropy": 4.994068193435669,
      "epoch": 2.474106247911794,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043850266277702446,
      "loss": 4.6488,
      "mean_token_accuracy": 0.22947092205286027,
      "num_tokens": 54500169.0,
      "step": 29620
    },
    {
      "entropy": 4.995990324020386,
      "epoch": 2.4745238890745074,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004384822100353155,
      "loss": 4.6502,
      "mean_token_accuracy": 0.2310274913907051,
      "num_tokens": 54508973.0,
      "step": 29625
    },
    {
      "entropy": 5.069343757629395,
      "epoch": 2.47494153023722,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004384617544316537,
      "loss": 4.7656,
      "mean_token_accuracy": 0.22161640375852584,
      "num_tokens": 54518450.0,
      "step": 29630
    },
    {
      "entropy": 5.039130020141601,
      "epoch": 2.475359171399933,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043844129596639714,
      "loss": 4.6876,
      "mean_token_accuracy": 0.22357116788625717,
      "num_tokens": 54527986.0,
      "step": 29635
    },
    {
      "entropy": 4.96745080947876,
      "epoch": 2.4757768125626463,
      "grad_norm": 1.140625,
      "learning_rate": 0.000438420834639904,
      "loss": 4.6193,
      "mean_token_accuracy": 0.23346778005361557,
      "num_tokens": 54536792.0,
      "step": 29640
    },
    {
      "entropy": 5.093621349334716,
      "epoch": 2.476194453725359,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00043840037045253234,
      "loss": 4.8937,
      "mean_token_accuracy": 0.2129406988620758,
      "num_tokens": 54547225.0,
      "step": 29645
    },
    {
      "entropy": 5.127464485168457,
      "epoch": 2.4766120948880723,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004383799034046405,
      "loss": 4.8262,
      "mean_token_accuracy": 0.21911160200834273,
      "num_tokens": 54557732.0,
      "step": 29650
    },
    {
      "entropy": 5.085922384262085,
      "epoch": 2.477029736050785,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004383594334965869,
      "loss": 4.7521,
      "mean_token_accuracy": 0.22126395851373673,
      "num_tokens": 54567510.0,
      "step": 29655
    },
    {
      "entropy": 5.088171291351318,
      "epoch": 2.4774473772134984,
      "grad_norm": 1.0,
      "learning_rate": 0.0004383389607287297,
      "loss": 4.7704,
      "mean_token_accuracy": 0.22680097371339797,
      "num_tokens": 54577619.0,
      "step": 29660
    },
    {
      "entropy": 4.946626329421997,
      "epoch": 2.477865018376211,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004383184851014274,
      "loss": 4.6289,
      "mean_token_accuracy": 0.22990991324186325,
      "num_tokens": 54586752.0,
      "step": 29665
    },
    {
      "entropy": 5.008446073532104,
      "epoch": 2.478282659538924,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043829800661503843,
      "loss": 4.7513,
      "mean_token_accuracy": 0.22466717213392257,
      "num_tokens": 54595757.0,
      "step": 29670
    },
    {
      "entropy": 5.079212236404419,
      "epoch": 2.4787003007016373,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043827752526992125,
      "loss": 4.762,
      "mean_token_accuracy": 0.2270022839307785,
      "num_tokens": 54605134.0,
      "step": 29675
    },
    {
      "entropy": 4.987422752380371,
      "epoch": 2.47911794186435,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00043825704106643454,
      "loss": 4.6007,
      "mean_token_accuracy": 0.23223561942577362,
      "num_tokens": 54614471.0,
      "step": 29680
    },
    {
      "entropy": 5.0161090850830075,
      "epoch": 2.4795355830270633,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043823655400493676,
      "loss": 4.7337,
      "mean_token_accuracy": 0.22045753449201583,
      "num_tokens": 54623466.0,
      "step": 29685
    },
    {
      "entropy": 5.0218347072601315,
      "epoch": 2.479953224189776,
      "grad_norm": 1.125,
      "learning_rate": 0.00043821606408578665,
      "loss": 4.6779,
      "mean_token_accuracy": 0.22985118329524995,
      "num_tokens": 54632898.0,
      "step": 29690
    },
    {
      "entropy": 5.094556570053101,
      "epoch": 2.480370865352489,
      "grad_norm": 1.0,
      "learning_rate": 0.0004381955713093428,
      "loss": 4.7456,
      "mean_token_accuracy": 0.2224603459239006,
      "num_tokens": 54642934.0,
      "step": 29695
    },
    {
      "entropy": 5.0010419368743895,
      "epoch": 2.480788506515202,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043817507567596415,
      "loss": 4.6748,
      "mean_token_accuracy": 0.22699926495552064,
      "num_tokens": 54652602.0,
      "step": 29700
    },
    {
      "entropy": 5.088768386840821,
      "epoch": 2.481206147677915,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043815457718600933,
      "loss": 4.7764,
      "mean_token_accuracy": 0.21917681246995926,
      "num_tokens": 54661801.0,
      "step": 29705
    },
    {
      "entropy": 5.088303852081299,
      "epoch": 2.4816237888406283,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004381340758398372,
      "loss": 4.7431,
      "mean_token_accuracy": 0.22738307267427443,
      "num_tokens": 54671977.0,
      "step": 29710
    },
    {
      "entropy": 5.007958316802979,
      "epoch": 2.482041430003341,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00043811357163780676,
      "loss": 4.6224,
      "mean_token_accuracy": 0.23441094905138016,
      "num_tokens": 54681820.0,
      "step": 29715
    },
    {
      "entropy": 5.009997797012329,
      "epoch": 2.482459071166054,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004380930645802769,
      "loss": 4.7172,
      "mean_token_accuracy": 0.22520692050457,
      "num_tokens": 54690713.0,
      "step": 29720
    },
    {
      "entropy": 5.092656946182251,
      "epoch": 2.482876712328767,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004380725546676067,
      "loss": 4.778,
      "mean_token_accuracy": 0.2152564600110054,
      "num_tokens": 54700808.0,
      "step": 29725
    },
    {
      "entropy": 4.963820219039917,
      "epoch": 2.48329435349148,
      "grad_norm": 1.125,
      "learning_rate": 0.00043805204190015503,
      "loss": 4.6417,
      "mean_token_accuracy": 0.234478659927845,
      "num_tokens": 54708415.0,
      "step": 29730
    },
    {
      "entropy": 4.914329290390015,
      "epoch": 2.4837119946541932,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004380315262782811,
      "loss": 4.6744,
      "mean_token_accuracy": 0.2217442125082016,
      "num_tokens": 54716887.0,
      "step": 29735
    },
    {
      "entropy": 5.102906513214111,
      "epoch": 2.484129635816906,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004380110078023441,
      "loss": 4.8001,
      "mean_token_accuracy": 0.21217101514339448,
      "num_tokens": 54727103.0,
      "step": 29740
    },
    {
      "entropy": 4.978630304336548,
      "epoch": 2.4845472769796193,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004379904864727032,
      "loss": 4.6122,
      "mean_token_accuracy": 0.2304440602660179,
      "num_tokens": 54735405.0,
      "step": 29745
    },
    {
      "entropy": 5.058162021636963,
      "epoch": 2.484964918142332,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043796996228971766,
      "loss": 4.7832,
      "mean_token_accuracy": 0.2208144560456276,
      "num_tokens": 54745278.0,
      "step": 29750
    },
    {
      "entropy": 5.027195978164673,
      "epoch": 2.485382559305045,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004379494352537467,
      "loss": 4.7002,
      "mean_token_accuracy": 0.23138374388217925,
      "num_tokens": 54753632.0,
      "step": 29755
    },
    {
      "entropy": 5.091578531265259,
      "epoch": 2.485800200467758,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004379289053651498,
      "loss": 4.8006,
      "mean_token_accuracy": 0.21550832241773604,
      "num_tokens": 54763924.0,
      "step": 29760
    },
    {
      "entropy": 4.946027231216431,
      "epoch": 2.486217841630471,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004379083726242862,
      "loss": 4.588,
      "mean_token_accuracy": 0.23151924461126328,
      "num_tokens": 54772747.0,
      "step": 29765
    },
    {
      "entropy": 5.055497646331787,
      "epoch": 2.4866354827931842,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004378878370315156,
      "loss": 4.7807,
      "mean_token_accuracy": 0.2293938770890236,
      "num_tokens": 54782325.0,
      "step": 29770
    },
    {
      "entropy": 4.984134387969971,
      "epoch": 2.487053123955897,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004378672985871972,
      "loss": 4.6924,
      "mean_token_accuracy": 0.22650362849235534,
      "num_tokens": 54791359.0,
      "step": 29775
    },
    {
      "entropy": 4.965299034118653,
      "epoch": 2.4874707651186103,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043784675729169076,
      "loss": 4.7016,
      "mean_token_accuracy": 0.22823719680309296,
      "num_tokens": 54800281.0,
      "step": 29780
    },
    {
      "entropy": 4.980224752426148,
      "epoch": 2.487888406281323,
      "grad_norm": 1.125,
      "learning_rate": 0.0004378262131453558,
      "loss": 4.6238,
      "mean_token_accuracy": 0.22761904150247575,
      "num_tokens": 54808954.0,
      "step": 29785
    },
    {
      "entropy": 5.068031740188599,
      "epoch": 2.488306047444036,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000437805666148552,
      "loss": 4.7139,
      "mean_token_accuracy": 0.23266490697860717,
      "num_tokens": 54817542.0,
      "step": 29790
    },
    {
      "entropy": 4.973852825164795,
      "epoch": 2.488723688606749,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043778511630163905,
      "loss": 4.6872,
      "mean_token_accuracy": 0.2256513699889183,
      "num_tokens": 54826189.0,
      "step": 29795
    },
    {
      "entropy": 5.091345024108887,
      "epoch": 2.489141329769462,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043776456360497665,
      "loss": 4.7285,
      "mean_token_accuracy": 0.22526886612176894,
      "num_tokens": 54835352.0,
      "step": 29800
    },
    {
      "entropy": 5.094847536087036,
      "epoch": 2.4895589709321753,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004377440080589247,
      "loss": 4.7396,
      "mean_token_accuracy": 0.21856761276721953,
      "num_tokens": 54844917.0,
      "step": 29805
    },
    {
      "entropy": 4.9913201332092285,
      "epoch": 2.489976612094888,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000437723449663843,
      "loss": 4.7674,
      "mean_token_accuracy": 0.21725212335586547,
      "num_tokens": 54854489.0,
      "step": 29810
    },
    {
      "entropy": 5.012646961212158,
      "epoch": 2.490394253257601,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004377028884200914,
      "loss": 4.7676,
      "mean_token_accuracy": 0.22502548694610597,
      "num_tokens": 54864462.0,
      "step": 29815
    },
    {
      "entropy": 5.014250040054321,
      "epoch": 2.490811894420314,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004376823243280298,
      "loss": 4.6827,
      "mean_token_accuracy": 0.22563070058822632,
      "num_tokens": 54873958.0,
      "step": 29820
    },
    {
      "entropy": 5.038641452789307,
      "epoch": 2.491229535583027,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004376617573880185,
      "loss": 4.7094,
      "mean_token_accuracy": 0.23188082575798036,
      "num_tokens": 54882553.0,
      "step": 29825
    },
    {
      "entropy": 5.063440036773682,
      "epoch": 2.49164717674574,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004376411876004172,
      "loss": 4.7211,
      "mean_token_accuracy": 0.22705531418323516,
      "num_tokens": 54891618.0,
      "step": 29830
    },
    {
      "entropy": 5.0626561641693115,
      "epoch": 2.492064817908453,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004376206149655862,
      "loss": 4.7708,
      "mean_token_accuracy": 0.23227239549160003,
      "num_tokens": 54900621.0,
      "step": 29835
    },
    {
      "entropy": 4.976263093948364,
      "epoch": 2.492482459071166,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004376000394838856,
      "loss": 4.6686,
      "mean_token_accuracy": 0.22938856184482576,
      "num_tokens": 54909116.0,
      "step": 29840
    },
    {
      "entropy": 5.061892652511597,
      "epoch": 2.492900100233879,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004375794611556755,
      "loss": 4.71,
      "mean_token_accuracy": 0.22601237297058105,
      "num_tokens": 54917940.0,
      "step": 29845
    },
    {
      "entropy": 4.991029262542725,
      "epoch": 2.493317741396592,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004375588799813163,
      "loss": 4.7776,
      "mean_token_accuracy": 0.22045097947120668,
      "num_tokens": 54927548.0,
      "step": 29850
    },
    {
      "entropy": 5.073188209533692,
      "epoch": 2.493735382559305,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004375382959611682,
      "loss": 4.7652,
      "mean_token_accuracy": 0.22431874722242356,
      "num_tokens": 54937157.0,
      "step": 29855
    },
    {
      "entropy": 5.112324047088623,
      "epoch": 2.494153023722018,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043751770909559155,
      "loss": 4.7225,
      "mean_token_accuracy": 0.22106694132089616,
      "num_tokens": 54946851.0,
      "step": 29860
    },
    {
      "entropy": 4.999168539047242,
      "epoch": 2.4945706648847312,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043749711938494693,
      "loss": 4.7735,
      "mean_token_accuracy": 0.22029085755348204,
      "num_tokens": 54956686.0,
      "step": 29865
    },
    {
      "entropy": 5.047651386260986,
      "epoch": 2.494988306047444,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043747652682959455,
      "loss": 4.6825,
      "mean_token_accuracy": 0.22513183504343032,
      "num_tokens": 54965233.0,
      "step": 29870
    },
    {
      "entropy": 5.091748285293579,
      "epoch": 2.495405947210157,
      "grad_norm": 1.125,
      "learning_rate": 0.0004374559314298949,
      "loss": 4.7155,
      "mean_token_accuracy": 0.22413330674171447,
      "num_tokens": 54973647.0,
      "step": 29875
    },
    {
      "entropy": 5.0510478019714355,
      "epoch": 2.49582358837287,
      "grad_norm": 1.046875,
      "learning_rate": 0.00043743533318620873,
      "loss": 4.7218,
      "mean_token_accuracy": 0.22067698389291762,
      "num_tokens": 54982550.0,
      "step": 29880
    },
    {
      "entropy": 4.948736763000488,
      "epoch": 2.496241229535583,
      "grad_norm": 1.0,
      "learning_rate": 0.00043741473209889654,
      "loss": 4.6065,
      "mean_token_accuracy": 0.230106720328331,
      "num_tokens": 54991137.0,
      "step": 29885
    },
    {
      "entropy": 4.96328592300415,
      "epoch": 2.496658870698296,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004373941281683189,
      "loss": 4.6896,
      "mean_token_accuracy": 0.22638812959194182,
      "num_tokens": 55000271.0,
      "step": 29890
    },
    {
      "entropy": 5.119955921173096,
      "epoch": 2.497076511861009,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004373735213948366,
      "loss": 4.8136,
      "mean_token_accuracy": 0.2173900008201599,
      "num_tokens": 55009379.0,
      "step": 29895
    },
    {
      "entropy": 5.026197290420532,
      "epoch": 2.4974941530237222,
      "grad_norm": 1.125,
      "learning_rate": 0.00043735291177881045,
      "loss": 4.6783,
      "mean_token_accuracy": 0.23555946946144105,
      "num_tokens": 55018562.0,
      "step": 29900
    },
    {
      "entropy": 5.047090816497803,
      "epoch": 2.497911794186435,
      "grad_norm": 1.125,
      "learning_rate": 0.000437332299320601,
      "loss": 4.744,
      "mean_token_accuracy": 0.22557324767112732,
      "num_tokens": 55027429.0,
      "step": 29905
    },
    {
      "entropy": 5.08401837348938,
      "epoch": 2.498329435349148,
      "grad_norm": 1.0,
      "learning_rate": 0.0004373116840205693,
      "loss": 4.7727,
      "mean_token_accuracy": 0.21747200042009354,
      "num_tokens": 55037510.0,
      "step": 29910
    },
    {
      "entropy": 5.063338708877564,
      "epoch": 2.498747076511861,
      "grad_norm": 1.234375,
      "learning_rate": 0.00043729106587907625,
      "loss": 4.6924,
      "mean_token_accuracy": 0.2326148569583893,
      "num_tokens": 55046944.0,
      "step": 29915
    },
    {
      "entropy": 5.008409070968628,
      "epoch": 2.499164717674574,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043727044489648263,
      "loss": 4.6613,
      "mean_token_accuracy": 0.23007450699806214,
      "num_tokens": 55056281.0,
      "step": 29920
    },
    {
      "entropy": 4.931233882904053,
      "epoch": 2.499582358837287,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004372498210731497,
      "loss": 4.6474,
      "mean_token_accuracy": 0.23152869790792466,
      "num_tokens": 55064764.0,
      "step": 29925
    },
    {
      "entropy": 5.028587388992309,
      "epoch": 2.5,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043722919440943833,
      "loss": 4.7199,
      "mean_token_accuracy": 0.22213349640369415,
      "num_tokens": 55074054.0,
      "step": 29930
    },
    {
      "entropy": 4.994639539718628,
      "epoch": 2.500417641162713,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004372085649057095,
      "loss": 4.6645,
      "mean_token_accuracy": 0.2341607391834259,
      "num_tokens": 55082623.0,
      "step": 29935
    },
    {
      "entropy": 5.102046728134155,
      "epoch": 2.500835282325426,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043718793256232463,
      "loss": 4.7799,
      "mean_token_accuracy": 0.22285906970500946,
      "num_tokens": 55092057.0,
      "step": 29940
    },
    {
      "entropy": 4.990512132644653,
      "epoch": 2.501252923488139,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043716729737964475,
      "loss": 4.6531,
      "mean_token_accuracy": 0.22880647331476212,
      "num_tokens": 55099887.0,
      "step": 29945
    },
    {
      "entropy": 5.045846366882325,
      "epoch": 2.501670564650852,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00043714665935803103,
      "loss": 4.6847,
      "mean_token_accuracy": 0.22571202516555786,
      "num_tokens": 55108143.0,
      "step": 29950
    },
    {
      "entropy": 5.045298385620117,
      "epoch": 2.502088205813565,
      "grad_norm": 1.046875,
      "learning_rate": 0.00043712601849784494,
      "loss": 4.7701,
      "mean_token_accuracy": 0.22143799513578416,
      "num_tokens": 55117920.0,
      "step": 29955
    },
    {
      "entropy": 5.165260887145996,
      "epoch": 2.5025058469762778,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043710537479944774,
      "loss": 4.8136,
      "mean_token_accuracy": 0.22521488815546037,
      "num_tokens": 55128557.0,
      "step": 29960
    },
    {
      "entropy": 4.963642024993897,
      "epoch": 2.502923488138991,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00043708472826320085,
      "loss": 4.6215,
      "mean_token_accuracy": 0.22662024050951005,
      "num_tokens": 55137886.0,
      "step": 29965
    },
    {
      "entropy": 5.068046379089355,
      "epoch": 2.503341129301704,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004370640788894656,
      "loss": 4.7904,
      "mean_token_accuracy": 0.2190541222691536,
      "num_tokens": 55146986.0,
      "step": 29970
    },
    {
      "entropy": 5.0340423583984375,
      "epoch": 2.503758770464417,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004370434266786037,
      "loss": 4.7034,
      "mean_token_accuracy": 0.22163422852754594,
      "num_tokens": 55156190.0,
      "step": 29975
    },
    {
      "entropy": 5.092615795135498,
      "epoch": 2.50417641162713,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004370227716309765,
      "loss": 4.6872,
      "mean_token_accuracy": 0.23222780972719193,
      "num_tokens": 55165591.0,
      "step": 29980
    },
    {
      "entropy": 5.0107439994812015,
      "epoch": 2.5045940527898427,
      "grad_norm": 1.015625,
      "learning_rate": 0.00043700211374694565,
      "loss": 4.6882,
      "mean_token_accuracy": 0.22298475354909897,
      "num_tokens": 55174865.0,
      "step": 29985
    },
    {
      "entropy": 5.0597114086151125,
      "epoch": 2.505011693952556,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004369814530268728,
      "loss": 4.7292,
      "mean_token_accuracy": 0.21899085640907287,
      "num_tokens": 55183199.0,
      "step": 29990
    },
    {
      "entropy": 5.134829711914063,
      "epoch": 2.505429335115269,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004369607894711196,
      "loss": 4.8068,
      "mean_token_accuracy": 0.21524416506290436,
      "num_tokens": 55192258.0,
      "step": 29995
    },
    {
      "entropy": 5.0541750431060795,
      "epoch": 2.505846976277982,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004369401230800479,
      "loss": 4.6938,
      "mean_token_accuracy": 0.23044013530015944,
      "num_tokens": 55200886.0,
      "step": 30000
    },
    {
      "epoch": 2.505846976277982,
      "eval_entropy": 4.826924989547196,
      "eval_loss": 4.7616190910339355,
      "eval_mean_token_accuracy": 0.23391976819038895,
      "eval_num_tokens": 55200886.0,
      "eval_runtime": 27.6074,
      "eval_samples_per_second": 1367.856,
      "eval_steps_per_second": 171.005,
      "step": 30000
    },
    {
      "entropy": 5.0630106925964355,
      "epoch": 2.506264617440695,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004369194538540194,
      "loss": 4.7167,
      "mean_token_accuracy": 0.220645073056221,
      "num_tokens": 55210553.0,
      "step": 30005
    },
    {
      "entropy": 5.032093572616577,
      "epoch": 2.506682258603408,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043689878179339593,
      "loss": 4.7529,
      "mean_token_accuracy": 0.2263756886124611,
      "num_tokens": 55219807.0,
      "step": 30010
    },
    {
      "entropy": 4.994658088684082,
      "epoch": 2.507099899766121,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004368781068985394,
      "loss": 4.6932,
      "mean_token_accuracy": 0.22247465699911118,
      "num_tokens": 55228970.0,
      "step": 30015
    },
    {
      "entropy": 5.060961580276489,
      "epoch": 2.507517540928834,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004368574291698118,
      "loss": 4.7365,
      "mean_token_accuracy": 0.2266113504767418,
      "num_tokens": 55238688.0,
      "step": 30020
    },
    {
      "entropy": 5.093770742416382,
      "epoch": 2.507935182091547,
      "grad_norm": 1.046875,
      "learning_rate": 0.000436836748607575,
      "loss": 4.684,
      "mean_token_accuracy": 0.22162377387285231,
      "num_tokens": 55247971.0,
      "step": 30025
    },
    {
      "entropy": 5.042075252532959,
      "epoch": 2.50835282325426,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004368160652121912,
      "loss": 4.7409,
      "mean_token_accuracy": 0.22922468036413193,
      "num_tokens": 55257806.0,
      "step": 30030
    },
    {
      "entropy": 5.0246641635894775,
      "epoch": 2.508770464416973,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004367953789840224,
      "loss": 4.7273,
      "mean_token_accuracy": 0.2301576927304268,
      "num_tokens": 55267339.0,
      "step": 30035
    },
    {
      "entropy": 5.023694181442261,
      "epoch": 2.509188105579686,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043677468992343074,
      "loss": 4.7719,
      "mean_token_accuracy": 0.222441965341568,
      "num_tokens": 55277374.0,
      "step": 30040
    },
    {
      "entropy": 5.139384603500366,
      "epoch": 2.509605746742399,
      "grad_norm": 1.125,
      "learning_rate": 0.00043675399803077837,
      "loss": 4.8208,
      "mean_token_accuracy": 0.21674130260944366,
      "num_tokens": 55286979.0,
      "step": 30045
    },
    {
      "entropy": 4.995148086547852,
      "epoch": 2.510023387905112,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004367333033064275,
      "loss": 4.6055,
      "mean_token_accuracy": 0.2307906687259674,
      "num_tokens": 55295999.0,
      "step": 30050
    },
    {
      "entropy": 5.009718656539917,
      "epoch": 2.5104410290678247,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004367126057507405,
      "loss": 4.7603,
      "mean_token_accuracy": 0.2211127758026123,
      "num_tokens": 55304977.0,
      "step": 30055
    },
    {
      "entropy": 5.093738603591919,
      "epoch": 2.510858670230538,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004366919053640797,
      "loss": 4.7255,
      "mean_token_accuracy": 0.22925444841384887,
      "num_tokens": 55314545.0,
      "step": 30060
    },
    {
      "entropy": 4.997061681747437,
      "epoch": 2.511276311393251,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004366712021468076,
      "loss": 4.6568,
      "mean_token_accuracy": 0.23196863383054733,
      "num_tokens": 55323546.0,
      "step": 30065
    },
    {
      "entropy": 4.946860885620117,
      "epoch": 2.511693952555964,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043665049609928634,
      "loss": 4.5578,
      "mean_token_accuracy": 0.23128837794065477,
      "num_tokens": 55332756.0,
      "step": 30070
    },
    {
      "entropy": 5.111238670349121,
      "epoch": 2.512111593718677,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004366297872218786,
      "loss": 4.8694,
      "mean_token_accuracy": 0.21423174887895585,
      "num_tokens": 55341864.0,
      "step": 30075
    },
    {
      "entropy": 4.999519395828247,
      "epoch": 2.5125292348813897,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00043660907551494696,
      "loss": 4.6796,
      "mean_token_accuracy": 0.22680197060108184,
      "num_tokens": 55351756.0,
      "step": 30080
    },
    {
      "entropy": 5.063880443572998,
      "epoch": 2.512946876044103,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043658836097885387,
      "loss": 4.7688,
      "mean_token_accuracy": 0.22219787836074828,
      "num_tokens": 55360959.0,
      "step": 30085
    },
    {
      "entropy": 5.084045362472534,
      "epoch": 2.5133645172068158,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043656764361396194,
      "loss": 4.8275,
      "mean_token_accuracy": 0.22179984599351882,
      "num_tokens": 55371026.0,
      "step": 30090
    },
    {
      "entropy": 5.0282210350036625,
      "epoch": 2.513782158369529,
      "grad_norm": 1.15625,
      "learning_rate": 0.000436546923420634,
      "loss": 4.6454,
      "mean_token_accuracy": 0.2279534935951233,
      "num_tokens": 55380091.0,
      "step": 30095
    },
    {
      "entropy": 4.972391843795776,
      "epoch": 2.514199799532242,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043652620039923276,
      "loss": 4.6488,
      "mean_token_accuracy": 0.22990714609622956,
      "num_tokens": 55388934.0,
      "step": 30100
    },
    {
      "entropy": 4.966113424301147,
      "epoch": 2.5146174406949546,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004365054745501209,
      "loss": 4.6648,
      "mean_token_accuracy": 0.22986818253993987,
      "num_tokens": 55397643.0,
      "step": 30105
    },
    {
      "entropy": 5.035957527160645,
      "epoch": 2.515035081857668,
      "grad_norm": 1.046875,
      "learning_rate": 0.00043648474587366134,
      "loss": 4.7559,
      "mean_token_accuracy": 0.2212809517979622,
      "num_tokens": 55406953.0,
      "step": 30110
    },
    {
      "entropy": 4.938359308242798,
      "epoch": 2.515452723020381,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004364640143702169,
      "loss": 4.5953,
      "mean_token_accuracy": 0.2320887789130211,
      "num_tokens": 55416623.0,
      "step": 30115
    },
    {
      "entropy": 5.163811445236206,
      "epoch": 2.515870364183094,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004364432800401505,
      "loss": 4.798,
      "mean_token_accuracy": 0.21971173584461212,
      "num_tokens": 55425818.0,
      "step": 30120
    },
    {
      "entropy": 5.173039197921753,
      "epoch": 2.5162880053458068,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043642254288382514,
      "loss": 4.7906,
      "mean_token_accuracy": 0.2121056064963341,
      "num_tokens": 55435104.0,
      "step": 30125
    },
    {
      "entropy": 4.955252170562744,
      "epoch": 2.51670564650852,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004364018029016039,
      "loss": 4.6211,
      "mean_token_accuracy": 0.23566503524780275,
      "num_tokens": 55443873.0,
      "step": 30130
    },
    {
      "entropy": 4.975338077545166,
      "epoch": 2.517123287671233,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004363810600938498,
      "loss": 4.7167,
      "mean_token_accuracy": 0.22904151529073716,
      "num_tokens": 55453780.0,
      "step": 30135
    },
    {
      "entropy": 5.056851196289062,
      "epoch": 2.517540928833946,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00043636031446092605,
      "loss": 4.8427,
      "mean_token_accuracy": 0.21754638254642486,
      "num_tokens": 55463648.0,
      "step": 30140
    },
    {
      "entropy": 5.081038570404052,
      "epoch": 2.517958569996659,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043633956600319575,
      "loss": 4.6809,
      "mean_token_accuracy": 0.22213131487369536,
      "num_tokens": 55473318.0,
      "step": 30145
    },
    {
      "entropy": 5.097248840332031,
      "epoch": 2.5183762111593717,
      "grad_norm": 1.015625,
      "learning_rate": 0.00043631881472102214,
      "loss": 4.8076,
      "mean_token_accuracy": 0.21542920172214508,
      "num_tokens": 55482511.0,
      "step": 30150
    },
    {
      "entropy": 5.048875617980957,
      "epoch": 2.518793852322085,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043629806061476847,
      "loss": 4.7231,
      "mean_token_accuracy": 0.22934584319591522,
      "num_tokens": 55491432.0,
      "step": 30155
    },
    {
      "entropy": 4.996987581253052,
      "epoch": 2.519211493484798,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004362773036847981,
      "loss": 4.7266,
      "mean_token_accuracy": 0.2298295632004738,
      "num_tokens": 55500825.0,
      "step": 30160
    },
    {
      "entropy": 5.079837369918823,
      "epoch": 2.519629134647511,
      "grad_norm": 1.0,
      "learning_rate": 0.00043625654393147445,
      "loss": 4.7384,
      "mean_token_accuracy": 0.22033980786800383,
      "num_tokens": 55509907.0,
      "step": 30165
    },
    {
      "entropy": 5.196114730834961,
      "epoch": 2.520046775810224,
      "grad_norm": 1.125,
      "learning_rate": 0.0004362357813551608,
      "loss": 4.872,
      "mean_token_accuracy": 0.21832302808761597,
      "num_tokens": 55519875.0,
      "step": 30170
    },
    {
      "entropy": 5.024299192428589,
      "epoch": 2.5204644169729367,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043621501595622077,
      "loss": 4.7027,
      "mean_token_accuracy": 0.22779758870601655,
      "num_tokens": 55529438.0,
      "step": 30175
    },
    {
      "entropy": 5.109128093719482,
      "epoch": 2.52088205813565,
      "grad_norm": 1.125,
      "learning_rate": 0.0004361942477350178,
      "loss": 4.7916,
      "mean_token_accuracy": 0.228518445789814,
      "num_tokens": 55537975.0,
      "step": 30180
    },
    {
      "entropy": 5.176058149337768,
      "epoch": 2.5212996992983627,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004361734766919156,
      "loss": 4.8569,
      "mean_token_accuracy": 0.21254616528749465,
      "num_tokens": 55547217.0,
      "step": 30185
    },
    {
      "entropy": 5.103442049026489,
      "epoch": 2.521717340461076,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004361527028272776,
      "loss": 4.7508,
      "mean_token_accuracy": 0.22403271943330766,
      "num_tokens": 55556050.0,
      "step": 30190
    },
    {
      "entropy": 5.034119462966919,
      "epoch": 2.522134981623789,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043613192614146753,
      "loss": 4.71,
      "mean_token_accuracy": 0.22969471514225007,
      "num_tokens": 55565393.0,
      "step": 30195
    },
    {
      "entropy": 4.991882562637329,
      "epoch": 2.5225526227865016,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004361111466348492,
      "loss": 4.6319,
      "mean_token_accuracy": 0.23450722247362138,
      "num_tokens": 55574060.0,
      "step": 30200
    },
    {
      "entropy": 5.064736986160279,
      "epoch": 2.522970263949215,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043609036430778623,
      "loss": 4.8048,
      "mean_token_accuracy": 0.21435664594173431,
      "num_tokens": 55583251.0,
      "step": 30205
    },
    {
      "entropy": 5.161922645568848,
      "epoch": 2.5233879051119277,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004360695791606426,
      "loss": 4.8513,
      "mean_token_accuracy": 0.21400541365146636,
      "num_tokens": 55593515.0,
      "step": 30210
    },
    {
      "entropy": 5.1429344654083256,
      "epoch": 2.523805546274641,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004360487911937821,
      "loss": 4.8052,
      "mean_token_accuracy": 0.22943765074014663,
      "num_tokens": 55602359.0,
      "step": 30215
    },
    {
      "entropy": 5.003453254699707,
      "epoch": 2.5242231874373537,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004360280004075687,
      "loss": 4.7047,
      "mean_token_accuracy": 0.23040456920862198,
      "num_tokens": 55611679.0,
      "step": 30220
    },
    {
      "entropy": 5.015108966827393,
      "epoch": 2.5246408286000666,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004360072068023662,
      "loss": 4.7585,
      "mean_token_accuracy": 0.23580467998981475,
      "num_tokens": 55621633.0,
      "step": 30225
    },
    {
      "entropy": 5.011137199401856,
      "epoch": 2.52505846976278,
      "grad_norm": 1.265625,
      "learning_rate": 0.00043598641037853885,
      "loss": 4.6918,
      "mean_token_accuracy": 0.22184576094150543,
      "num_tokens": 55629821.0,
      "step": 30230
    },
    {
      "entropy": 4.949507999420166,
      "epoch": 2.525476110925493,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043596561113645055,
      "loss": 4.5757,
      "mean_token_accuracy": 0.22503385543823243,
      "num_tokens": 55638898.0,
      "step": 30235
    },
    {
      "entropy": 4.961554098129272,
      "epoch": 2.525893752088206,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004359448090764655,
      "loss": 4.6873,
      "mean_token_accuracy": 0.22363023310899735,
      "num_tokens": 55648309.0,
      "step": 30240
    },
    {
      "entropy": 5.0981369972229,
      "epoch": 2.5263113932509187,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043592400419894785,
      "loss": 4.8281,
      "mean_token_accuracy": 0.21603368520736693,
      "num_tokens": 55657889.0,
      "step": 30245
    },
    {
      "entropy": 5.050623226165771,
      "epoch": 2.526729034413632,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004359031965042617,
      "loss": 4.7473,
      "mean_token_accuracy": 0.21898823827505112,
      "num_tokens": 55666743.0,
      "step": 30250
    },
    {
      "entropy": 4.9955791473388675,
      "epoch": 2.5271466755763448,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00043588238599277156,
      "loss": 4.5844,
      "mean_token_accuracy": 0.23316707611083984,
      "num_tokens": 55675821.0,
      "step": 30255
    },
    {
      "entropy": 5.113960838317871,
      "epoch": 2.527564316739058,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00043586157266484155,
      "loss": 4.7567,
      "mean_token_accuracy": 0.21560672670602798,
      "num_tokens": 55685623.0,
      "step": 30260
    },
    {
      "entropy": 5.02050666809082,
      "epoch": 2.527981957901771,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004358407565208361,
      "loss": 4.6125,
      "mean_token_accuracy": 0.2390730857849121,
      "num_tokens": 55694017.0,
      "step": 30265
    },
    {
      "entropy": 4.9762077808380125,
      "epoch": 2.5283995990644836,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043581993756111956,
      "loss": 4.7107,
      "mean_token_accuracy": 0.22733474969863893,
      "num_tokens": 55703685.0,
      "step": 30270
    },
    {
      "entropy": 4.963444328308105,
      "epoch": 2.528817240227197,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004357991157860565,
      "loss": 4.6225,
      "mean_token_accuracy": 0.2386380597949028,
      "num_tokens": 55711555.0,
      "step": 30275
    },
    {
      "entropy": 5.019357824325562,
      "epoch": 2.5292348813899097,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004357782911960113,
      "loss": 4.7066,
      "mean_token_accuracy": 0.22213375121355056,
      "num_tokens": 55720313.0,
      "step": 30280
    },
    {
      "entropy": 4.9933630466461185,
      "epoch": 2.529652522552623,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004357574637913486,
      "loss": 4.7405,
      "mean_token_accuracy": 0.22208440452814102,
      "num_tokens": 55729092.0,
      "step": 30285
    },
    {
      "entropy": 5.021745920181274,
      "epoch": 2.5300701637153358,
      "grad_norm": 1.15625,
      "learning_rate": 0.000435736633572433,
      "loss": 4.6407,
      "mean_token_accuracy": 0.23189745098352432,
      "num_tokens": 55738843.0,
      "step": 30290
    },
    {
      "entropy": 5.087292861938477,
      "epoch": 2.5304878048780486,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004357158005396291,
      "loss": 4.8211,
      "mean_token_accuracy": 0.2266951471567154,
      "num_tokens": 55749070.0,
      "step": 30295
    },
    {
      "entropy": 4.9575948238372805,
      "epoch": 2.530905446040762,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043569496469330173,
      "loss": 4.6649,
      "mean_token_accuracy": 0.23322119414806367,
      "num_tokens": 55758327.0,
      "step": 30300
    },
    {
      "entropy": 5.07608585357666,
      "epoch": 2.5313230872034747,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004356741260338156,
      "loss": 4.779,
      "mean_token_accuracy": 0.220491024851799,
      "num_tokens": 55767344.0,
      "step": 30305
    },
    {
      "entropy": 5.150390481948852,
      "epoch": 2.531740728366188,
      "grad_norm": 1.03125,
      "learning_rate": 0.00043565328456153534,
      "loss": 4.8561,
      "mean_token_accuracy": 0.2202373445034027,
      "num_tokens": 55776905.0,
      "step": 30310
    },
    {
      "entropy": 5.078128910064697,
      "epoch": 2.5321583695289007,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004356324402768261,
      "loss": 4.7814,
      "mean_token_accuracy": 0.21775510907173157,
      "num_tokens": 55785442.0,
      "step": 30315
    },
    {
      "entropy": 5.025175952911377,
      "epoch": 2.5325760106916135,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004356115931800526,
      "loss": 4.7573,
      "mean_token_accuracy": 0.21972140222787856,
      "num_tokens": 55794917.0,
      "step": 30320
    },
    {
      "entropy": 5.014276599884033,
      "epoch": 2.532993651854327,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004355907432715798,
      "loss": 4.6426,
      "mean_token_accuracy": 0.2347886562347412,
      "num_tokens": 55803249.0,
      "step": 30325
    },
    {
      "entropy": 4.92359881401062,
      "epoch": 2.5334112930170396,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043556989055177275,
      "loss": 4.6267,
      "mean_token_accuracy": 0.23219176828861238,
      "num_tokens": 55811604.0,
      "step": 30330
    },
    {
      "entropy": 4.978626394271851,
      "epoch": 2.533828934179753,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043554903502099644,
      "loss": 4.6431,
      "mean_token_accuracy": 0.23484944701194763,
      "num_tokens": 55820818.0,
      "step": 30335
    },
    {
      "entropy": 5.03688907623291,
      "epoch": 2.5342465753424657,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043552817667961616,
      "loss": 4.7199,
      "mean_token_accuracy": 0.2312193274497986,
      "num_tokens": 55830171.0,
      "step": 30340
    },
    {
      "entropy": 4.992161989212036,
      "epoch": 2.5346642165051785,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043550731552799676,
      "loss": 4.6047,
      "mean_token_accuracy": 0.2347714811563492,
      "num_tokens": 55838748.0,
      "step": 30345
    },
    {
      "entropy": 4.94439468383789,
      "epoch": 2.5350818576678917,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043548645156650376,
      "loss": 4.6822,
      "mean_token_accuracy": 0.22655651867389678,
      "num_tokens": 55847672.0,
      "step": 30350
    },
    {
      "entropy": 5.103990077972412,
      "epoch": 2.535499498830605,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004354655847955021,
      "loss": 4.7606,
      "mean_token_accuracy": 0.2146633580327034,
      "num_tokens": 55856822.0,
      "step": 30355
    },
    {
      "entropy": 5.108082962036133,
      "epoch": 2.535917139993318,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043544471521535726,
      "loss": 4.7592,
      "mean_token_accuracy": 0.2212297037243843,
      "num_tokens": 55867463.0,
      "step": 30360
    },
    {
      "entropy": 5.087140464782715,
      "epoch": 2.5363347811560306,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043542384282643467,
      "loss": 4.8522,
      "mean_token_accuracy": 0.2105454072356224,
      "num_tokens": 55877108.0,
      "step": 30365
    },
    {
      "entropy": 5.025195932388305,
      "epoch": 2.536752422318744,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004354029676290995,
      "loss": 4.6188,
      "mean_token_accuracy": 0.2289562225341797,
      "num_tokens": 55886504.0,
      "step": 30370
    },
    {
      "entropy": 5.022847414016724,
      "epoch": 2.5371700634814567,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004353820896237173,
      "loss": 4.654,
      "mean_token_accuracy": 0.22874696105718612,
      "num_tokens": 55895296.0,
      "step": 30375
    },
    {
      "entropy": 5.008752918243408,
      "epoch": 2.53758770464417,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004353612088106536,
      "loss": 4.7415,
      "mean_token_accuracy": 0.22459916472435,
      "num_tokens": 55904151.0,
      "step": 30380
    },
    {
      "entropy": 4.979145526885986,
      "epoch": 2.5380053458068828,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043534032519027395,
      "loss": 4.7175,
      "mean_token_accuracy": 0.2405565083026886,
      "num_tokens": 55912667.0,
      "step": 30385
    },
    {
      "entropy": 5.094192552566528,
      "epoch": 2.5384229869695956,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004353194387629438,
      "loss": 4.8138,
      "mean_token_accuracy": 0.22602093517780303,
      "num_tokens": 55922647.0,
      "step": 30390
    },
    {
      "entropy": 4.991261959075928,
      "epoch": 2.538840628132309,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000435298549529029,
      "loss": 4.7146,
      "mean_token_accuracy": 0.22978295385837555,
      "num_tokens": 55932040.0,
      "step": 30395
    },
    {
      "entropy": 5.122245502471924,
      "epoch": 2.5392582692950216,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043527765748889506,
      "loss": 4.8419,
      "mean_token_accuracy": 0.21866804957389832,
      "num_tokens": 55940882.0,
      "step": 30400
    },
    {
      "entropy": 4.942188692092896,
      "epoch": 2.539675910457735,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004352567626429079,
      "loss": 4.6001,
      "mean_token_accuracy": 0.22968150824308395,
      "num_tokens": 55949522.0,
      "step": 30405
    },
    {
      "entropy": 4.934399509429932,
      "epoch": 2.5400935516204477,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004352358649914331,
      "loss": 4.6237,
      "mean_token_accuracy": 0.23459719866514206,
      "num_tokens": 55957529.0,
      "step": 30410
    },
    {
      "entropy": 5.066319465637207,
      "epoch": 2.5405111927831605,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043521496453483667,
      "loss": 4.7026,
      "mean_token_accuracy": 0.22512824088335037,
      "num_tokens": 55967400.0,
      "step": 30415
    },
    {
      "entropy": 5.100576734542846,
      "epoch": 2.5409288339458738,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043519406127348435,
      "loss": 4.7243,
      "mean_token_accuracy": 0.23369622081518174,
      "num_tokens": 55976262.0,
      "step": 30420
    },
    {
      "entropy": 4.986144590377807,
      "epoch": 2.5413464751085866,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004351731552077423,
      "loss": 4.7059,
      "mean_token_accuracy": 0.22388728260993956,
      "num_tokens": 55985751.0,
      "step": 30425
    },
    {
      "entropy": 5.032533597946167,
      "epoch": 2.5417641162713,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004351522463379763,
      "loss": 4.7354,
      "mean_token_accuracy": 0.2220069110393524,
      "num_tokens": 55995365.0,
      "step": 30430
    },
    {
      "entropy": 5.030676794052124,
      "epoch": 2.5421817574340126,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004351313346645525,
      "loss": 4.6958,
      "mean_token_accuracy": 0.2260548084974289,
      "num_tokens": 56004271.0,
      "step": 30435
    },
    {
      "entropy": 4.978684616088867,
      "epoch": 2.5425993985967255,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004351104201878368,
      "loss": 4.6923,
      "mean_token_accuracy": 0.22918299883604049,
      "num_tokens": 56013326.0,
      "step": 30440
    },
    {
      "entropy": 5.016396760940552,
      "epoch": 2.5430170397594387,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004350895029081956,
      "loss": 4.6947,
      "mean_token_accuracy": 0.22989387065172195,
      "num_tokens": 56021664.0,
      "step": 30445
    },
    {
      "entropy": 5.051750612258911,
      "epoch": 2.5434346809221515,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00043506858282599487,
      "loss": 4.7091,
      "mean_token_accuracy": 0.2272404208779335,
      "num_tokens": 56030030.0,
      "step": 30450
    },
    {
      "entropy": 5.01701021194458,
      "epoch": 2.543852322084865,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000435047659941601,
      "loss": 4.7296,
      "mean_token_accuracy": 0.22356920540332795,
      "num_tokens": 56039939.0,
      "step": 30455
    },
    {
      "entropy": 4.987038660049438,
      "epoch": 2.5442699632475776,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043502673425538016,
      "loss": 4.6647,
      "mean_token_accuracy": 0.22884376645088195,
      "num_tokens": 56048534.0,
      "step": 30460
    },
    {
      "entropy": 4.98088436126709,
      "epoch": 2.5446876044102904,
      "grad_norm": 1.125,
      "learning_rate": 0.00043500580576769874,
      "loss": 4.6671,
      "mean_token_accuracy": 0.23348767012357713,
      "num_tokens": 56057242.0,
      "step": 30465
    },
    {
      "entropy": 5.021131610870361,
      "epoch": 2.5451052455730037,
      "grad_norm": 1.15625,
      "learning_rate": 0.000434984874478923,
      "loss": 4.6755,
      "mean_token_accuracy": 0.22469871044158934,
      "num_tokens": 56065615.0,
      "step": 30470
    },
    {
      "entropy": 4.963023567199707,
      "epoch": 2.545522886735717,
      "grad_norm": 1.125,
      "learning_rate": 0.00043496394038941956,
      "loss": 4.6132,
      "mean_token_accuracy": 0.23356594294309616,
      "num_tokens": 56074351.0,
      "step": 30475
    },
    {
      "entropy": 4.978484201431274,
      "epoch": 2.5459405278984297,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00043494300349955473,
      "loss": 4.6324,
      "mean_token_accuracy": 0.23066116124391556,
      "num_tokens": 56083915.0,
      "step": 30480
    },
    {
      "entropy": 5.079688358306885,
      "epoch": 2.5463581690611425,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004349220638096952,
      "loss": 4.7806,
      "mean_token_accuracy": 0.21319733262062074,
      "num_tokens": 56093424.0,
      "step": 30485
    },
    {
      "entropy": 5.076727771759034,
      "epoch": 2.546775810223856,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004349011213202073,
      "loss": 4.7875,
      "mean_token_accuracy": 0.22403163611888885,
      "num_tokens": 56103798.0,
      "step": 30490
    },
    {
      "entropy": 5.102654266357422,
      "epoch": 2.5471934513865686,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00043488017603145787,
      "loss": 4.7807,
      "mean_token_accuracy": 0.21915305703878402,
      "num_tokens": 56113841.0,
      "step": 30495
    },
    {
      "entropy": 5.106951761245727,
      "epoch": 2.547611092549282,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043485922794381357,
      "loss": 4.8056,
      "mean_token_accuracy": 0.21666458398103713,
      "num_tokens": 56123294.0,
      "step": 30500
    },
    {
      "entropy": 5.019442892074585,
      "epoch": 2.5480287337119947,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00043483827705764094,
      "loss": 4.6631,
      "mean_token_accuracy": 0.22926327884197234,
      "num_tokens": 56133041.0,
      "step": 30505
    },
    {
      "entropy": 5.018097352981568,
      "epoch": 2.5484463748747075,
      "grad_norm": 1.03125,
      "learning_rate": 0.00043481732337330697,
      "loss": 4.784,
      "mean_token_accuracy": 0.22600394487380981,
      "num_tokens": 56142796.0,
      "step": 30510
    },
    {
      "entropy": 5.078743362426758,
      "epoch": 2.5488640160374207,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043479636689117834,
      "loss": 4.7431,
      "mean_token_accuracy": 0.2193443015217781,
      "num_tokens": 56152912.0,
      "step": 30515
    },
    {
      "entropy": 5.042441940307617,
      "epoch": 2.5492816572001336,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043477540761162196,
      "loss": 4.7256,
      "mean_token_accuracy": 0.22184675484895705,
      "num_tokens": 56161860.0,
      "step": 30520
    },
    {
      "entropy": 4.964613151550293,
      "epoch": 2.549699298362847,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004347544455350047,
      "loss": 4.7647,
      "mean_token_accuracy": 0.23000606894493103,
      "num_tokens": 56171525.0,
      "step": 30525
    },
    {
      "entropy": 5.082145929336548,
      "epoch": 2.5501169395255596,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043473348066169363,
      "loss": 4.7062,
      "mean_token_accuracy": 0.21972536444664,
      "num_tokens": 56181219.0,
      "step": 30530
    },
    {
      "entropy": 5.049448251724243,
      "epoch": 2.5505345806882724,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004347125129920557,
      "loss": 4.6226,
      "mean_token_accuracy": 0.23670080602169036,
      "num_tokens": 56189983.0,
      "step": 30535
    },
    {
      "entropy": 5.001484632492065,
      "epoch": 2.5509522218509857,
      "grad_norm": 1.25,
      "learning_rate": 0.0004346915425264581,
      "loss": 4.7082,
      "mean_token_accuracy": 0.2246992379426956,
      "num_tokens": 56199150.0,
      "step": 30540
    },
    {
      "entropy": 4.9954026222229,
      "epoch": 2.5513698630136985,
      "grad_norm": 1.125,
      "learning_rate": 0.0004346705692652677,
      "loss": 4.6959,
      "mean_token_accuracy": 0.22281949669122697,
      "num_tokens": 56208550.0,
      "step": 30545
    },
    {
      "entropy": 4.9366936683654785,
      "epoch": 2.5517875041764118,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004346495932088518,
      "loss": 4.5784,
      "mean_token_accuracy": 0.23690212666988372,
      "num_tokens": 56217487.0,
      "step": 30550
    },
    {
      "entropy": 5.013050556182861,
      "epoch": 2.5522051453391246,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004346286143575776,
      "loss": 4.7474,
      "mean_token_accuracy": 0.22132721543312073,
      "num_tokens": 56226969.0,
      "step": 30555
    },
    {
      "entropy": 5.092245721817017,
      "epoch": 2.5526227865018374,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043460763271181235,
      "loss": 4.8512,
      "mean_token_accuracy": 0.21823169589042662,
      "num_tokens": 56236223.0,
      "step": 30560
    },
    {
      "entropy": 5.009548473358154,
      "epoch": 2.5530404276645506,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043458664827192333,
      "loss": 4.6172,
      "mean_token_accuracy": 0.23901292234659194,
      "num_tokens": 56246319.0,
      "step": 30565
    },
    {
      "entropy": 5.0989405632019045,
      "epoch": 2.5534580688272634,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004345656610382781,
      "loss": 4.8739,
      "mean_token_accuracy": 0.21652993261814119,
      "num_tokens": 56256270.0,
      "step": 30570
    },
    {
      "entropy": 5.065709638595581,
      "epoch": 2.5538757099899767,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004345446710112438,
      "loss": 4.7336,
      "mean_token_accuracy": 0.2222206026315689,
      "num_tokens": 56265454.0,
      "step": 30575
    },
    {
      "entropy": 5.024135780334473,
      "epoch": 2.5542933511526895,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043452367819118797,
      "loss": 4.7738,
      "mean_token_accuracy": 0.2273286208510399,
      "num_tokens": 56274986.0,
      "step": 30580
    },
    {
      "entropy": 5.02663779258728,
      "epoch": 2.5547109923154023,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043450268257847814,
      "loss": 4.7464,
      "mean_token_accuracy": 0.22288685739040376,
      "num_tokens": 56284608.0,
      "step": 30585
    },
    {
      "entropy": 4.973872423171997,
      "epoch": 2.5551286334781156,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004344816841734819,
      "loss": 4.64,
      "mean_token_accuracy": 0.23455039113759996,
      "num_tokens": 56292508.0,
      "step": 30590
    },
    {
      "entropy": 5.045935010910034,
      "epoch": 2.555546274640829,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004344606829765668,
      "loss": 4.7443,
      "mean_token_accuracy": 0.22296129763126374,
      "num_tokens": 56302338.0,
      "step": 30595
    },
    {
      "entropy": 5.041334772109986,
      "epoch": 2.5559639158035417,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004344396789881005,
      "loss": 4.7181,
      "mean_token_accuracy": 0.22633254826068877,
      "num_tokens": 56311147.0,
      "step": 30600
    },
    {
      "entropy": 5.0238665580749515,
      "epoch": 2.5563815569662545,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004344186722084507,
      "loss": 4.7245,
      "mean_token_accuracy": 0.231918603181839,
      "num_tokens": 56320614.0,
      "step": 30605
    },
    {
      "entropy": 4.995738124847412,
      "epoch": 2.5567991981289677,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043439766263798506,
      "loss": 4.7629,
      "mean_token_accuracy": 0.22272588610649108,
      "num_tokens": 56330411.0,
      "step": 30610
    },
    {
      "entropy": 5.101046085357666,
      "epoch": 2.5572168392916805,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043437665027707155,
      "loss": 4.739,
      "mean_token_accuracy": 0.226634581387043,
      "num_tokens": 56339644.0,
      "step": 30615
    },
    {
      "entropy": 5.064119911193847,
      "epoch": 2.557634480454394,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043435563512607796,
      "loss": 4.7026,
      "mean_token_accuracy": 0.22285614013671876,
      "num_tokens": 56348533.0,
      "step": 30620
    },
    {
      "entropy": 5.09104905128479,
      "epoch": 2.5580521216171066,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004343346171853722,
      "loss": 4.7054,
      "mean_token_accuracy": 0.22249774038791656,
      "num_tokens": 56358533.0,
      "step": 30625
    },
    {
      "entropy": 4.997280216217041,
      "epoch": 2.5584697627798194,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004343135964553221,
      "loss": 4.7213,
      "mean_token_accuracy": 0.22824340760707856,
      "num_tokens": 56367739.0,
      "step": 30630
    },
    {
      "entropy": 5.039799976348877,
      "epoch": 2.5588874039425327,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004342925729362957,
      "loss": 4.6388,
      "mean_token_accuracy": 0.22688556611537933,
      "num_tokens": 56375315.0,
      "step": 30635
    },
    {
      "entropy": 5.024776840209961,
      "epoch": 2.5593050451052455,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004342715466286611,
      "loss": 4.6788,
      "mean_token_accuracy": 0.23679903149604797,
      "num_tokens": 56384226.0,
      "step": 30640
    },
    {
      "entropy": 5.055185556411743,
      "epoch": 2.5597226862679587,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043425051753278633,
      "loss": 4.6995,
      "mean_token_accuracy": 0.22952677607536315,
      "num_tokens": 56393249.0,
      "step": 30645
    },
    {
      "entropy": 5.088347387313843,
      "epoch": 2.5601403274306715,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004342294856490397,
      "loss": 4.7737,
      "mean_token_accuracy": 0.21872366964817047,
      "num_tokens": 56402701.0,
      "step": 30650
    },
    {
      "entropy": 5.008662176132202,
      "epoch": 2.5605579685933844,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043420845097778913,
      "loss": 4.6737,
      "mean_token_accuracy": 0.2273968443274498,
      "num_tokens": 56411927.0,
      "step": 30655
    },
    {
      "entropy": 5.110662984848022,
      "epoch": 2.5609756097560976,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000434187413519403,
      "loss": 4.8718,
      "mean_token_accuracy": 0.21900838762521743,
      "num_tokens": 56421852.0,
      "step": 30660
    },
    {
      "entropy": 5.066917276382446,
      "epoch": 2.5613932509188104,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043416637327424955,
      "loss": 4.6779,
      "mean_token_accuracy": 0.2263822540640831,
      "num_tokens": 56431428.0,
      "step": 30665
    },
    {
      "entropy": 5.000027465820312,
      "epoch": 2.5618108920815237,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004341453302426972,
      "loss": 4.6383,
      "mean_token_accuracy": 0.23714437931776047,
      "num_tokens": 56440339.0,
      "step": 30670
    },
    {
      "entropy": 5.033366537094116,
      "epoch": 2.5622285332442365,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004341242844251143,
      "loss": 4.7615,
      "mean_token_accuracy": 0.22593214511871337,
      "num_tokens": 56450407.0,
      "step": 30675
    },
    {
      "entropy": 5.0885748863220215,
      "epoch": 2.5626461744069493,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004341032358218692,
      "loss": 4.7352,
      "mean_token_accuracy": 0.22401726245880127,
      "num_tokens": 56460357.0,
      "step": 30680
    },
    {
      "entropy": 5.107408332824707,
      "epoch": 2.5630638155696626,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043408218443333056,
      "loss": 4.722,
      "mean_token_accuracy": 0.2227082997560501,
      "num_tokens": 56469433.0,
      "step": 30685
    },
    {
      "entropy": 4.955031204223633,
      "epoch": 2.563481456732376,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043406113025986666,
      "loss": 4.6534,
      "mean_token_accuracy": 0.22897257506847382,
      "num_tokens": 56478608.0,
      "step": 30690
    },
    {
      "entropy": 4.948844528198242,
      "epoch": 2.5638990978950886,
      "grad_norm": 1.125,
      "learning_rate": 0.0004340400733018463,
      "loss": 4.6408,
      "mean_token_accuracy": 0.23301138430833818,
      "num_tokens": 56487904.0,
      "step": 30695
    },
    {
      "entropy": 5.074830722808838,
      "epoch": 2.5643167390578014,
      "grad_norm": 1.03125,
      "learning_rate": 0.000434019013559638,
      "loss": 4.8349,
      "mean_token_accuracy": 0.21982724964618683,
      "num_tokens": 56498126.0,
      "step": 30700
    },
    {
      "entropy": 4.9977658748626705,
      "epoch": 2.5647343802205147,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043399795103361046,
      "loss": 4.6279,
      "mean_token_accuracy": 0.2326365351676941,
      "num_tokens": 56506400.0,
      "step": 30705
    },
    {
      "entropy": 5.010205411911011,
      "epoch": 2.5651520213832275,
      "grad_norm": 1.03125,
      "learning_rate": 0.00043397688572413236,
      "loss": 4.7277,
      "mean_token_accuracy": 0.21951897889375688,
      "num_tokens": 56516458.0,
      "step": 30710
    },
    {
      "entropy": 4.888632822036743,
      "epoch": 2.5655696625459408,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043395581763157256,
      "loss": 4.6122,
      "mean_token_accuracy": 0.2347693473100662,
      "num_tokens": 56524586.0,
      "step": 30715
    },
    {
      "entropy": 5.038855314254761,
      "epoch": 2.5659873037086536,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043393474675629983,
      "loss": 4.7129,
      "mean_token_accuracy": 0.22262288630008698,
      "num_tokens": 56533797.0,
      "step": 30720
    },
    {
      "entropy": 5.08210597038269,
      "epoch": 2.5664049448713664,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043391367309868304,
      "loss": 4.7171,
      "mean_token_accuracy": 0.22446295171976088,
      "num_tokens": 56543976.0,
      "step": 30725
    },
    {
      "entropy": 5.084735107421875,
      "epoch": 2.5668225860340796,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00043389259665909107,
      "loss": 4.721,
      "mean_token_accuracy": 0.21770584136247634,
      "num_tokens": 56553421.0,
      "step": 30730
    },
    {
      "entropy": 5.0419464111328125,
      "epoch": 2.5672402271967925,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000433871517437893,
      "loss": 4.8178,
      "mean_token_accuracy": 0.21603239625692366,
      "num_tokens": 56562780.0,
      "step": 30735
    },
    {
      "entropy": 4.966608762741089,
      "epoch": 2.5676578683595057,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043385043543545777,
      "loss": 4.6438,
      "mean_token_accuracy": 0.2276611343026161,
      "num_tokens": 56571186.0,
      "step": 30740
    },
    {
      "entropy": 4.975108194351196,
      "epoch": 2.5680755095222185,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043382935065215444,
      "loss": 4.6668,
      "mean_token_accuracy": 0.23300876915454866,
      "num_tokens": 56579282.0,
      "step": 30745
    },
    {
      "entropy": 5.005797672271728,
      "epoch": 2.5684931506849313,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043380826308835226,
      "loss": 4.7035,
      "mean_token_accuracy": 0.23001219630241393,
      "num_tokens": 56587962.0,
      "step": 30750
    },
    {
      "entropy": 5.051454353332519,
      "epoch": 2.5689107918476446,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043378717274442015,
      "loss": 4.8016,
      "mean_token_accuracy": 0.22208006531000138,
      "num_tokens": 56598075.0,
      "step": 30755
    },
    {
      "entropy": 5.052069282531738,
      "epoch": 2.5693284330103574,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043376607962072747,
      "loss": 4.7185,
      "mean_token_accuracy": 0.22444116771221162,
      "num_tokens": 56607363.0,
      "step": 30760
    },
    {
      "entropy": 5.031454610824585,
      "epoch": 2.5697460741730707,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043374498371764344,
      "loss": 4.6614,
      "mean_token_accuracy": 0.23525035828351976,
      "num_tokens": 56616041.0,
      "step": 30765
    },
    {
      "entropy": 5.0848081588745115,
      "epoch": 2.5701637153357835,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043372388503553743,
      "loss": 4.7221,
      "mean_token_accuracy": 0.22271555215120314,
      "num_tokens": 56625901.0,
      "step": 30770
    },
    {
      "entropy": 4.958749580383301,
      "epoch": 2.5705813564984963,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00043370278357477874,
      "loss": 4.7065,
      "mean_token_accuracy": 0.22949864119291305,
      "num_tokens": 56636255.0,
      "step": 30775
    },
    {
      "entropy": 4.9808272361755375,
      "epoch": 2.5709989976612095,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043368167933573675,
      "loss": 4.6711,
      "mean_token_accuracy": 0.23210528194904329,
      "num_tokens": 56645045.0,
      "step": 30780
    },
    {
      "entropy": 5.034532737731934,
      "epoch": 2.5714166388239224,
      "grad_norm": 1.1875,
      "learning_rate": 0.000433660572318781,
      "loss": 4.6855,
      "mean_token_accuracy": 0.23714314997196198,
      "num_tokens": 56653063.0,
      "step": 30785
    },
    {
      "entropy": 4.972627305984497,
      "epoch": 2.5718342799866356,
      "grad_norm": 1.015625,
      "learning_rate": 0.000433639462524281,
      "loss": 4.6781,
      "mean_token_accuracy": 0.22380623519420623,
      "num_tokens": 56663060.0,
      "step": 30790
    },
    {
      "entropy": 4.959523773193359,
      "epoch": 2.5722519211493484,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004336183499526062,
      "loss": 4.6278,
      "mean_token_accuracy": 0.23209887593984604,
      "num_tokens": 56672134.0,
      "step": 30795
    },
    {
      "entropy": 5.026049613952637,
      "epoch": 2.5726695623120612,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043359723460412625,
      "loss": 4.6955,
      "mean_token_accuracy": 0.2282773271203041,
      "num_tokens": 56681182.0,
      "step": 30800
    },
    {
      "entropy": 5.0022117614746096,
      "epoch": 2.5730872034747745,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004335761164792108,
      "loss": 4.5769,
      "mean_token_accuracy": 0.23958368003368377,
      "num_tokens": 56689375.0,
      "step": 30805
    },
    {
      "entropy": 5.026850414276123,
      "epoch": 2.5735048446374877,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004335549955782296,
      "loss": 4.7887,
      "mean_token_accuracy": 0.2217339515686035,
      "num_tokens": 56699190.0,
      "step": 30810
    },
    {
      "entropy": 5.042739582061768,
      "epoch": 2.5739224858002006,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004335338719015523,
      "loss": 4.7212,
      "mean_token_accuracy": 0.23325949311256408,
      "num_tokens": 56708332.0,
      "step": 30815
    },
    {
      "entropy": 5.155475091934204,
      "epoch": 2.5743401269629134,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004335127454495488,
      "loss": 4.7744,
      "mean_token_accuracy": 0.2246915504336357,
      "num_tokens": 56718163.0,
      "step": 30820
    },
    {
      "entropy": 5.041335153579712,
      "epoch": 2.5747577681256266,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004334916162225889,
      "loss": 4.6947,
      "mean_token_accuracy": 0.2261391893029213,
      "num_tokens": 56727976.0,
      "step": 30825
    },
    {
      "entropy": 5.0385767936706545,
      "epoch": 2.5751754092883394,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004334704842210424,
      "loss": 4.7825,
      "mean_token_accuracy": 0.2234594151377678,
      "num_tokens": 56738161.0,
      "step": 30830
    },
    {
      "entropy": 5.063410329818725,
      "epoch": 2.5755930504510527,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004334493494452794,
      "loss": 4.7829,
      "mean_token_accuracy": 0.22068548053503037,
      "num_tokens": 56747672.0,
      "step": 30835
    },
    {
      "entropy": 4.9977470397949215,
      "epoch": 2.5760106916137655,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004334282118956697,
      "loss": 4.6444,
      "mean_token_accuracy": 0.2341902494430542,
      "num_tokens": 56756923.0,
      "step": 30840
    },
    {
      "entropy": 5.060463809967041,
      "epoch": 2.5764283327764783,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004334070715725836,
      "loss": 4.7434,
      "mean_token_accuracy": 0.22702683210372926,
      "num_tokens": 56766030.0,
      "step": 30845
    },
    {
      "entropy": 4.974031162261963,
      "epoch": 2.5768459739391916,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00043338592847639093,
      "loss": 4.6596,
      "mean_token_accuracy": 0.22047845423221588,
      "num_tokens": 56775208.0,
      "step": 30850
    },
    {
      "entropy": 5.12632155418396,
      "epoch": 2.5772636151019044,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004333647826074619,
      "loss": 4.8319,
      "mean_token_accuracy": 0.22285894751548768,
      "num_tokens": 56784221.0,
      "step": 30855
    },
    {
      "entropy": 5.013577890396118,
      "epoch": 2.5776812562646176,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004333436339661668,
      "loss": 4.6201,
      "mean_token_accuracy": 0.23346790820360183,
      "num_tokens": 56792089.0,
      "step": 30860
    },
    {
      "entropy": 4.942790508270264,
      "epoch": 2.5780988974273304,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043332248255287584,
      "loss": 4.7432,
      "mean_token_accuracy": 0.2230277955532074,
      "num_tokens": 56802346.0,
      "step": 30865
    },
    {
      "entropy": 5.072631120681763,
      "epoch": 2.5785165385900433,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043330132836795916,
      "loss": 4.7026,
      "mean_token_accuracy": 0.22699021697044372,
      "num_tokens": 56811326.0,
      "step": 30870
    },
    {
      "entropy": 4.99421067237854,
      "epoch": 2.5789341797527565,
      "grad_norm": 1.125,
      "learning_rate": 0.00043328017141178724,
      "loss": 4.5671,
      "mean_token_accuracy": 0.24593627452850342,
      "num_tokens": 56820382.0,
      "step": 30875
    },
    {
      "entropy": 4.9646869659423825,
      "epoch": 2.5793518209154693,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043325901168473035,
      "loss": 4.6583,
      "mean_token_accuracy": 0.22666958272457122,
      "num_tokens": 56830678.0,
      "step": 30880
    },
    {
      "entropy": 4.9802172660827635,
      "epoch": 2.5797694620781826,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043323784918715896,
      "loss": 4.6896,
      "mean_token_accuracy": 0.22921402901411056,
      "num_tokens": 56840457.0,
      "step": 30885
    },
    {
      "entropy": 5.021861743927002,
      "epoch": 2.5801871032408954,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043321668391944364,
      "loss": 4.6976,
      "mean_token_accuracy": 0.22371830344200133,
      "num_tokens": 56849360.0,
      "step": 30890
    },
    {
      "entropy": 5.106096887588501,
      "epoch": 2.580604744403608,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004331955158819547,
      "loss": 4.8955,
      "mean_token_accuracy": 0.21244917511940004,
      "num_tokens": 56858514.0,
      "step": 30895
    },
    {
      "entropy": 5.07537841796875,
      "epoch": 2.5810223855663215,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043317434507506285,
      "loss": 4.7386,
      "mean_token_accuracy": 0.23000474125146866,
      "num_tokens": 56867800.0,
      "step": 30900
    },
    {
      "entropy": 5.05389256477356,
      "epoch": 2.5814400267290343,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043315317149913876,
      "loss": 4.6832,
      "mean_token_accuracy": 0.2217813327908516,
      "num_tokens": 56876777.0,
      "step": 30905
    },
    {
      "entropy": 5.094754934310913,
      "epoch": 2.5818576678917475,
      "grad_norm": 1.046875,
      "learning_rate": 0.000433131995154553,
      "loss": 4.7725,
      "mean_token_accuracy": 0.2301408603787422,
      "num_tokens": 56885937.0,
      "step": 30910
    },
    {
      "entropy": 5.0553131103515625,
      "epoch": 2.5822753090544603,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004331108160416764,
      "loss": 4.716,
      "mean_token_accuracy": 0.22258265763521196,
      "num_tokens": 56895340.0,
      "step": 30915
    },
    {
      "entropy": 4.950269174575806,
      "epoch": 2.582692950217173,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004330896341608795,
      "loss": 4.6733,
      "mean_token_accuracy": 0.2327854409813881,
      "num_tokens": 56903975.0,
      "step": 30920
    },
    {
      "entropy": 5.042997550964356,
      "epoch": 2.5831105913798864,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043306844951253336,
      "loss": 4.7071,
      "mean_token_accuracy": 0.22440023720264435,
      "num_tokens": 56913396.0,
      "step": 30925
    },
    {
      "entropy": 5.0288807392120365,
      "epoch": 2.5835282325425997,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043304726209700874,
      "loss": 4.7122,
      "mean_token_accuracy": 0.23081065714359283,
      "num_tokens": 56923161.0,
      "step": 30930
    },
    {
      "entropy": 5.03877329826355,
      "epoch": 2.5839458737053125,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004330260719146765,
      "loss": 4.702,
      "mean_token_accuracy": 0.2310839682817459,
      "num_tokens": 56932449.0,
      "step": 30935
    },
    {
      "entropy": 4.997699642181397,
      "epoch": 2.5843635148680253,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004330048789659077,
      "loss": 4.7319,
      "mean_token_accuracy": 0.22880424410104752,
      "num_tokens": 56941146.0,
      "step": 30940
    },
    {
      "entropy": 4.956087446212768,
      "epoch": 2.5847811560307385,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043298368325107335,
      "loss": 4.5891,
      "mean_token_accuracy": 0.2388443633913994,
      "num_tokens": 56949909.0,
      "step": 30945
    },
    {
      "entropy": 5.022794246673584,
      "epoch": 2.5851987971934514,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004329624847705444,
      "loss": 4.7245,
      "mean_token_accuracy": 0.22533673495054246,
      "num_tokens": 56960291.0,
      "step": 30950
    },
    {
      "entropy": 5.069640588760376,
      "epoch": 2.5856164383561646,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043294128352469195,
      "loss": 4.8331,
      "mean_token_accuracy": 0.21282802522182465,
      "num_tokens": 56970490.0,
      "step": 30955
    },
    {
      "entropy": 5.045568466186523,
      "epoch": 2.5860340795188774,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00043292007951388733,
      "loss": 4.7551,
      "mean_token_accuracy": 0.22613574415445328,
      "num_tokens": 56979763.0,
      "step": 30960
    },
    {
      "entropy": 5.05817723274231,
      "epoch": 2.5864517206815902,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00043289887273850167,
      "loss": 4.7513,
      "mean_token_accuracy": 0.2268397569656372,
      "num_tokens": 56989932.0,
      "step": 30965
    },
    {
      "entropy": 5.080754184722901,
      "epoch": 2.5868693618443035,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000432877663198906,
      "loss": 4.686,
      "mean_token_accuracy": 0.22618281245231628,
      "num_tokens": 56999263.0,
      "step": 30970
    },
    {
      "entropy": 5.05094895362854,
      "epoch": 2.5872870030070163,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004328564508954719,
      "loss": 4.794,
      "mean_token_accuracy": 0.21937690675258636,
      "num_tokens": 57009243.0,
      "step": 30975
    },
    {
      "entropy": 4.92084527015686,
      "epoch": 2.5877046441697296,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004328352358285706,
      "loss": 4.593,
      "mean_token_accuracy": 0.23428644835948945,
      "num_tokens": 57018154.0,
      "step": 30980
    },
    {
      "entropy": 5.163352203369141,
      "epoch": 2.5881222853324424,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043281401799857356,
      "loss": 4.8148,
      "mean_token_accuracy": 0.21562607884407042,
      "num_tokens": 57027237.0,
      "step": 30985
    },
    {
      "entropy": 5.079004430770874,
      "epoch": 2.588539926495155,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004327927974058521,
      "loss": 4.67,
      "mean_token_accuracy": 0.22352788001298904,
      "num_tokens": 57036426.0,
      "step": 30990
    },
    {
      "entropy": 5.0095627784729,
      "epoch": 2.5889575676578684,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004327715740507778,
      "loss": 4.7371,
      "mean_token_accuracy": 0.22761392444372178,
      "num_tokens": 57044233.0,
      "step": 30995
    },
    {
      "entropy": 5.005588674545288,
      "epoch": 2.5893752088205813,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004327503479337222,
      "loss": 4.825,
      "mean_token_accuracy": 0.21378623694181442,
      "num_tokens": 57054222.0,
      "step": 31000
    },
    {
      "entropy": 5.119524192810059,
      "epoch": 2.5897928499832945,
      "grad_norm": 1.125,
      "learning_rate": 0.00043272911905505696,
      "loss": 4.81,
      "mean_token_accuracy": 0.21992631554603576,
      "num_tokens": 57062990.0,
      "step": 31005
    },
    {
      "entropy": 5.0747971534729,
      "epoch": 2.5902104911460073,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004327078874151535,
      "loss": 4.6952,
      "mean_token_accuracy": 0.22233896255493163,
      "num_tokens": 57071769.0,
      "step": 31010
    },
    {
      "entropy": 5.075671911239624,
      "epoch": 2.59062813230872,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004326866530143838,
      "loss": 4.7202,
      "mean_token_accuracy": 0.22853921949863434,
      "num_tokens": 57080171.0,
      "step": 31015
    },
    {
      "entropy": 5.078958559036255,
      "epoch": 2.5910457734714334,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043266541585311924,
      "loss": 4.7258,
      "mean_token_accuracy": 0.23035914450883865,
      "num_tokens": 57089241.0,
      "step": 31020
    },
    {
      "entropy": 5.146998119354248,
      "epoch": 2.591463414634146,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004326441759317319,
      "loss": 4.7611,
      "mean_token_accuracy": 0.22510810941457748,
      "num_tokens": 57098037.0,
      "step": 31025
    },
    {
      "entropy": 5.001893758773804,
      "epoch": 2.5918810557968595,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004326229332505934,
      "loss": 4.7568,
      "mean_token_accuracy": 0.22051631212234496,
      "num_tokens": 57108322.0,
      "step": 31030
    },
    {
      "entropy": 4.980502939224243,
      "epoch": 2.5922986969595723,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004326016878100759,
      "loss": 4.6197,
      "mean_token_accuracy": 0.22779424637556075,
      "num_tokens": 57117131.0,
      "step": 31035
    },
    {
      "entropy": 5.076739120483398,
      "epoch": 2.592716338122285,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00043258043961055104,
      "loss": 4.66,
      "mean_token_accuracy": 0.2270619288086891,
      "num_tokens": 57126278.0,
      "step": 31040
    },
    {
      "entropy": 5.072905349731445,
      "epoch": 2.5931339792849983,
      "grad_norm": 1.046875,
      "learning_rate": 0.00043255918865239097,
      "loss": 4.7324,
      "mean_token_accuracy": 0.23115535974502563,
      "num_tokens": 57135163.0,
      "step": 31045
    },
    {
      "entropy": 5.007380676269531,
      "epoch": 2.5935516204477116,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043253793493596755,
      "loss": 4.7076,
      "mean_token_accuracy": 0.2207816421985626,
      "num_tokens": 57144368.0,
      "step": 31050
    },
    {
      "entropy": 4.937808179855347,
      "epoch": 2.5939692616104244,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000432516678461653,
      "loss": 4.6431,
      "mean_token_accuracy": 0.22657214105129242,
      "num_tokens": 57153421.0,
      "step": 31055
    },
    {
      "entropy": 5.053151178359985,
      "epoch": 2.594386902773137,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004324954192298194,
      "loss": 4.7588,
      "mean_token_accuracy": 0.2166458860039711,
      "num_tokens": 57163238.0,
      "step": 31060
    },
    {
      "entropy": 5.120047473907471,
      "epoch": 2.5948045439358505,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043247415724083886,
      "loss": 4.7902,
      "mean_token_accuracy": 0.22049783319234847,
      "num_tokens": 57172711.0,
      "step": 31065
    },
    {
      "entropy": 4.996895503997803,
      "epoch": 2.5952221850985633,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00043245289249508375,
      "loss": 4.6943,
      "mean_token_accuracy": 0.23653963059186936,
      "num_tokens": 57181841.0,
      "step": 31070
    },
    {
      "entropy": 4.9573646068573,
      "epoch": 2.5956398262612765,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004324316249929261,
      "loss": 4.7167,
      "mean_token_accuracy": 0.22795581817626953,
      "num_tokens": 57191195.0,
      "step": 31075
    },
    {
      "entropy": 4.9447790622711185,
      "epoch": 2.5960574674239894,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004324103547347384,
      "loss": 4.5894,
      "mean_token_accuracy": 0.23934667110443114,
      "num_tokens": 57200050.0,
      "step": 31080
    },
    {
      "entropy": 5.139016914367676,
      "epoch": 2.596475108586702,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043238908172089306,
      "loss": 4.7573,
      "mean_token_accuracy": 0.22706261575222014,
      "num_tokens": 57208296.0,
      "step": 31085
    },
    {
      "entropy": 4.949224376678467,
      "epoch": 2.5968927497494154,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043236780595176227,
      "loss": 4.6392,
      "mean_token_accuracy": 0.22115678787231446,
      "num_tokens": 57217658.0,
      "step": 31090
    },
    {
      "entropy": 4.999518728256225,
      "epoch": 2.5973103909121282,
      "grad_norm": 1.125,
      "learning_rate": 0.0004323465274277187,
      "loss": 4.6704,
      "mean_token_accuracy": 0.2264449864625931,
      "num_tokens": 57227573.0,
      "step": 31095
    },
    {
      "entropy": 4.986409282684326,
      "epoch": 2.5977280320748415,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004323252461491347,
      "loss": 4.648,
      "mean_token_accuracy": 0.23088318556547166,
      "num_tokens": 57236958.0,
      "step": 31100
    },
    {
      "entropy": 4.966069650650025,
      "epoch": 2.5981456732375543,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004323039621163829,
      "loss": 4.6886,
      "mean_token_accuracy": 0.22817837297916413,
      "num_tokens": 57246064.0,
      "step": 31105
    },
    {
      "entropy": 5.075095224380493,
      "epoch": 2.598563314400267,
      "grad_norm": 1.0625,
      "learning_rate": 0.000432282675329836,
      "loss": 4.7908,
      "mean_token_accuracy": 0.21754492670297623,
      "num_tokens": 57257162.0,
      "step": 31110
    },
    {
      "entropy": 5.112594747543335,
      "epoch": 2.5989809555629804,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043226138578986644,
      "loss": 4.7775,
      "mean_token_accuracy": 0.21806465834379196,
      "num_tokens": 57266080.0,
      "step": 31115
    },
    {
      "entropy": 5.094799137115478,
      "epoch": 2.599398596725693,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043224009349684706,
      "loss": 4.7456,
      "mean_token_accuracy": 0.2284609407186508,
      "num_tokens": 57274717.0,
      "step": 31120
    },
    {
      "entropy": 5.0274378776550295,
      "epoch": 2.5998162378884064,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004322187984511506,
      "loss": 4.7411,
      "mean_token_accuracy": 0.2165293276309967,
      "num_tokens": 57283691.0,
      "step": 31125
    },
    {
      "entropy": 4.97036828994751,
      "epoch": 2.6002338790511192,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004321975006531498,
      "loss": 4.6065,
      "mean_token_accuracy": 0.22493280172348024,
      "num_tokens": 57291393.0,
      "step": 31130
    },
    {
      "entropy": 4.954540109634399,
      "epoch": 2.600651520213832,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004321762001032175,
      "loss": 4.642,
      "mean_token_accuracy": 0.23567113280296326,
      "num_tokens": 57300188.0,
      "step": 31135
    },
    {
      "entropy": 5.090503931045532,
      "epoch": 2.6010691613765453,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043215489680172665,
      "loss": 4.8386,
      "mean_token_accuracy": 0.212135049700737,
      "num_tokens": 57309306.0,
      "step": 31140
    },
    {
      "entropy": 5.140288066864014,
      "epoch": 2.601486802539258,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004321335907490502,
      "loss": 4.7249,
      "mean_token_accuracy": 0.22482967227697373,
      "num_tokens": 57319149.0,
      "step": 31145
    },
    {
      "entropy": 4.995416069030762,
      "epoch": 2.6019044437019714,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004321122819455611,
      "loss": 4.7693,
      "mean_token_accuracy": 0.22296308130025863,
      "num_tokens": 57328766.0,
      "step": 31150
    },
    {
      "entropy": 5.104329919815063,
      "epoch": 2.602322084864684,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043209097039163233,
      "loss": 4.7558,
      "mean_token_accuracy": 0.21634989976882935,
      "num_tokens": 57338231.0,
      "step": 31155
    },
    {
      "entropy": 4.988959789276123,
      "epoch": 2.602739726027397,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043206965608763706,
      "loss": 4.6569,
      "mean_token_accuracy": 0.23095210492610932,
      "num_tokens": 57347307.0,
      "step": 31160
    },
    {
      "entropy": 5.056548118591309,
      "epoch": 2.6031573671901103,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004320483390339484,
      "loss": 4.756,
      "mean_token_accuracy": 0.23137522041797637,
      "num_tokens": 57356143.0,
      "step": 31165
    },
    {
      "entropy": 5.062110042572021,
      "epoch": 2.6035750083528235,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004320270192309395,
      "loss": 4.7473,
      "mean_token_accuracy": 0.22078770101070405,
      "num_tokens": 57366283.0,
      "step": 31170
    },
    {
      "entropy": 5.02216477394104,
      "epoch": 2.6039926495155363,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004320056966789836,
      "loss": 4.7164,
      "mean_token_accuracy": 0.22746148258447646,
      "num_tokens": 57375781.0,
      "step": 31175
    },
    {
      "entropy": 5.025474262237549,
      "epoch": 2.604410290678249,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043198437137845403,
      "loss": 4.6613,
      "mean_token_accuracy": 0.23413978964090348,
      "num_tokens": 57384851.0,
      "step": 31180
    },
    {
      "entropy": 5.075723886489868,
      "epoch": 2.6048279318409624,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043196304332972414,
      "loss": 4.7511,
      "mean_token_accuracy": 0.22023862451314927,
      "num_tokens": 57393843.0,
      "step": 31185
    },
    {
      "entropy": 5.012979173660279,
      "epoch": 2.605245573003675,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043194171253316716,
      "loss": 4.7186,
      "mean_token_accuracy": 0.22833411991596222,
      "num_tokens": 57403248.0,
      "step": 31190
    },
    {
      "entropy": 5.038263654708862,
      "epoch": 2.6056632141663885,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043192037898915665,
      "loss": 4.6588,
      "mean_token_accuracy": 0.23073093444108964,
      "num_tokens": 57412626.0,
      "step": 31195
    },
    {
      "entropy": 5.0534008026123045,
      "epoch": 2.6060808553291013,
      "grad_norm": 1.078125,
      "learning_rate": 0.000431899042698066,
      "loss": 4.7093,
      "mean_token_accuracy": 0.2244963154196739,
      "num_tokens": 57421747.0,
      "step": 31200
    },
    {
      "entropy": 5.170133590698242,
      "epoch": 2.606498496491814,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004318777036602687,
      "loss": 4.8909,
      "mean_token_accuracy": 0.20954708606004716,
      "num_tokens": 57431106.0,
      "step": 31205
    },
    {
      "entropy": 5.017666053771973,
      "epoch": 2.6069161376545273,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043185636187613837,
      "loss": 4.6084,
      "mean_token_accuracy": 0.23174284249544144,
      "num_tokens": 57440210.0,
      "step": 31210
    },
    {
      "entropy": 5.0471296310424805,
      "epoch": 2.60733377881724,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004318350173460487,
      "loss": 4.7434,
      "mean_token_accuracy": 0.22132954001426697,
      "num_tokens": 57449012.0,
      "step": 31215
    },
    {
      "entropy": 5.019529676437378,
      "epoch": 2.6077514199799534,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004318136700703731,
      "loss": 4.7141,
      "mean_token_accuracy": 0.22243162244558334,
      "num_tokens": 57457143.0,
      "step": 31220
    },
    {
      "entropy": 5.007430553436279,
      "epoch": 2.6081690611426662,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004317923200494856,
      "loss": 4.6338,
      "mean_token_accuracy": 0.23724062442779542,
      "num_tokens": 57465791.0,
      "step": 31225
    },
    {
      "entropy": 4.992727136611938,
      "epoch": 2.608586702305379,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043177096728375976,
      "loss": 4.6996,
      "mean_token_accuracy": 0.2339504912495613,
      "num_tokens": 57474749.0,
      "step": 31230
    },
    {
      "entropy": 5.048696851730346,
      "epoch": 2.6090043434680923,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043174961177356946,
      "loss": 4.7302,
      "mean_token_accuracy": 0.22437763661146165,
      "num_tokens": 57484895.0,
      "step": 31235
    },
    {
      "entropy": 5.046771574020386,
      "epoch": 2.609421984630805,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004317282535192885,
      "loss": 4.6572,
      "mean_token_accuracy": 0.23704418838024138,
      "num_tokens": 57493751.0,
      "step": 31240
    },
    {
      "entropy": 4.9450318813323975,
      "epoch": 2.6098396257935184,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00043170689252129083,
      "loss": 4.6121,
      "mean_token_accuracy": 0.24144421368837357,
      "num_tokens": 57503096.0,
      "step": 31245
    },
    {
      "entropy": 4.954338407516479,
      "epoch": 2.610257266956231,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004316855287799504,
      "loss": 4.6204,
      "mean_token_accuracy": 0.23588238060474395,
      "num_tokens": 57512667.0,
      "step": 31250
    },
    {
      "entropy": 4.9932207584381105,
      "epoch": 2.610674908118944,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043166416229564107,
      "loss": 4.7531,
      "mean_token_accuracy": 0.22401953637599945,
      "num_tokens": 57522404.0,
      "step": 31255
    },
    {
      "entropy": 4.946794748306274,
      "epoch": 2.6110925492816572,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00043164279306873713,
      "loss": 4.5669,
      "mean_token_accuracy": 0.23261688202619552,
      "num_tokens": 57531213.0,
      "step": 31260
    },
    {
      "entropy": 5.092269468307495,
      "epoch": 2.61151019044437,
      "grad_norm": 1.125,
      "learning_rate": 0.00043162142109961246,
      "loss": 4.8111,
      "mean_token_accuracy": 0.2137394741177559,
      "num_tokens": 57540943.0,
      "step": 31265
    },
    {
      "entropy": 5.091990375518799,
      "epoch": 2.6119278316070833,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004316000463886414,
      "loss": 4.7563,
      "mean_token_accuracy": 0.21899642050266266,
      "num_tokens": 57550054.0,
      "step": 31270
    },
    {
      "entropy": 5.134695816040039,
      "epoch": 2.612345472769796,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004315786689361979,
      "loss": 4.7759,
      "mean_token_accuracy": 0.22235514372587203,
      "num_tokens": 57559683.0,
      "step": 31275
    },
    {
      "entropy": 5.013331937789917,
      "epoch": 2.612763113932509,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043155728874265636,
      "loss": 4.7487,
      "mean_token_accuracy": 0.22181136906147003,
      "num_tokens": 57569446.0,
      "step": 31280
    },
    {
      "entropy": 5.004734373092651,
      "epoch": 2.613180755095222,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00043153590580839106,
      "loss": 4.6597,
      "mean_token_accuracy": 0.23411049842834472,
      "num_tokens": 57578468.0,
      "step": 31285
    },
    {
      "entropy": 5.063075542449951,
      "epoch": 2.6135983962579354,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043151452013377636,
      "loss": 4.7466,
      "mean_token_accuracy": 0.2256305992603302,
      "num_tokens": 57587017.0,
      "step": 31290
    },
    {
      "entropy": 5.12561182975769,
      "epoch": 2.6140160374206483,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043149313171918647,
      "loss": 4.8433,
      "mean_token_accuracy": 0.22007427662611007,
      "num_tokens": 57596688.0,
      "step": 31295
    },
    {
      "entropy": 5.044176054000855,
      "epoch": 2.614433678583361,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043147174056499605,
      "loss": 4.6864,
      "mean_token_accuracy": 0.22746388465166092,
      "num_tokens": 57606595.0,
      "step": 31300
    },
    {
      "entropy": 5.090537881851196,
      "epoch": 2.6148513197460743,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00043145034667157935,
      "loss": 4.7844,
      "mean_token_accuracy": 0.22203228920698165,
      "num_tokens": 57616413.0,
      "step": 31305
    },
    {
      "entropy": 5.198417282104492,
      "epoch": 2.615268960908787,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004314289500393111,
      "loss": 4.7855,
      "mean_token_accuracy": 0.227092881500721,
      "num_tokens": 57625903.0,
      "step": 31310
    },
    {
      "entropy": 4.95264778137207,
      "epoch": 2.6156866020715004,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043140755066856573,
      "loss": 4.6547,
      "mean_token_accuracy": 0.22936769425868989,
      "num_tokens": 57635315.0,
      "step": 31315
    },
    {
      "entropy": 4.9319243907928465,
      "epoch": 2.616104243234213,
      "grad_norm": 1.109375,
      "learning_rate": 0.000431386148559718,
      "loss": 4.6758,
      "mean_token_accuracy": 0.23864946216344834,
      "num_tokens": 57643806.0,
      "step": 31320
    },
    {
      "entropy": 4.9915399074554445,
      "epoch": 2.616521884396926,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043136474371314245,
      "loss": 4.6111,
      "mean_token_accuracy": 0.2242615208029747,
      "num_tokens": 57654030.0,
      "step": 31325
    },
    {
      "entropy": 5.0706109523773195,
      "epoch": 2.6169395255596393,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004313433361292138,
      "loss": 4.6032,
      "mean_token_accuracy": 0.24318066686391832,
      "num_tokens": 57663141.0,
      "step": 31330
    },
    {
      "entropy": 5.065397453308106,
      "epoch": 2.617357166722352,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043132192580830694,
      "loss": 4.8226,
      "mean_token_accuracy": 0.2244413822889328,
      "num_tokens": 57672943.0,
      "step": 31335
    },
    {
      "entropy": 5.002715682983398,
      "epoch": 2.6177748078850653,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043130051275079654,
      "loss": 4.6695,
      "mean_token_accuracy": 0.22801540344953536,
      "num_tokens": 57682086.0,
      "step": 31340
    },
    {
      "entropy": 5.050452470779419,
      "epoch": 2.618192449047778,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004312790969570575,
      "loss": 4.7231,
      "mean_token_accuracy": 0.22522563934326173,
      "num_tokens": 57691229.0,
      "step": 31345
    },
    {
      "entropy": 5.063477563858032,
      "epoch": 2.618610090210491,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004312576784274648,
      "loss": 4.7259,
      "mean_token_accuracy": 0.2317351758480072,
      "num_tokens": 57700057.0,
      "step": 31350
    },
    {
      "entropy": 5.048080968856811,
      "epoch": 2.619027731373204,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004312362571623934,
      "loss": 4.6846,
      "mean_token_accuracy": 0.22372951358556747,
      "num_tokens": 57709498.0,
      "step": 31355
    },
    {
      "entropy": 5.029263734817505,
      "epoch": 2.619445372535917,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004312148331622182,
      "loss": 4.727,
      "mean_token_accuracy": 0.2317760020494461,
      "num_tokens": 57718719.0,
      "step": 31360
    },
    {
      "entropy": 5.0181896686553955,
      "epoch": 2.6198630136986303,
      "grad_norm": 1.0,
      "learning_rate": 0.0004311934064273143,
      "loss": 4.6075,
      "mean_token_accuracy": 0.23383424878120423,
      "num_tokens": 57727225.0,
      "step": 31365
    },
    {
      "entropy": 4.972310876846313,
      "epoch": 2.620280654861343,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004311719769580569,
      "loss": 4.6227,
      "mean_token_accuracy": 0.238717982172966,
      "num_tokens": 57736027.0,
      "step": 31370
    },
    {
      "entropy": 4.98313717842102,
      "epoch": 2.620698296024056,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00043115054475482094,
      "loss": 4.6336,
      "mean_token_accuracy": 0.2308762103319168,
      "num_tokens": 57745456.0,
      "step": 31375
    },
    {
      "entropy": 5.018719434738159,
      "epoch": 2.621115937186769,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004311291098179818,
      "loss": 4.743,
      "mean_token_accuracy": 0.22120047360658646,
      "num_tokens": 57754689.0,
      "step": 31380
    },
    {
      "entropy": 4.976802968978882,
      "epoch": 2.621533578349482,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043110767214791466,
      "loss": 4.6559,
      "mean_token_accuracy": 0.22292463332414628,
      "num_tokens": 57763759.0,
      "step": 31385
    },
    {
      "entropy": 5.0395369052886965,
      "epoch": 2.6219512195121952,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004310862317449948,
      "loss": 4.644,
      "mean_token_accuracy": 0.23828192800283432,
      "num_tokens": 57772765.0,
      "step": 31390
    },
    {
      "entropy": 5.06029634475708,
      "epoch": 2.622368860674908,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004310647886095976,
      "loss": 4.7468,
      "mean_token_accuracy": 0.2199746251106262,
      "num_tokens": 57783831.0,
      "step": 31395
    },
    {
      "entropy": 5.043554210662842,
      "epoch": 2.622786501837621,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004310433427420984,
      "loss": 4.6708,
      "mean_token_accuracy": 0.23278746008872986,
      "num_tokens": 57792439.0,
      "step": 31400
    },
    {
      "entropy": 4.995661687850952,
      "epoch": 2.623204143000334,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043102189414287255,
      "loss": 4.7128,
      "mean_token_accuracy": 0.22685018330812454,
      "num_tokens": 57802280.0,
      "step": 31405
    },
    {
      "entropy": 5.038001203536988,
      "epoch": 2.6236217841630474,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004310004428122958,
      "loss": 4.7191,
      "mean_token_accuracy": 0.23384272307157516,
      "num_tokens": 57811631.0,
      "step": 31410
    },
    {
      "entropy": 5.074638032913208,
      "epoch": 2.62403942532576,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043097898875074344,
      "loss": 4.7126,
      "mean_token_accuracy": 0.22505296915769576,
      "num_tokens": 57821286.0,
      "step": 31415
    },
    {
      "entropy": 5.045477437973022,
      "epoch": 2.624457066488473,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004309575319585912,
      "loss": 4.6514,
      "mean_token_accuracy": 0.22863131016492844,
      "num_tokens": 57830446.0,
      "step": 31420
    },
    {
      "entropy": 4.982852268218994,
      "epoch": 2.6248747076511862,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043093607243621456,
      "loss": 4.6262,
      "mean_token_accuracy": 0.23752336353063583,
      "num_tokens": 57839203.0,
      "step": 31425
    },
    {
      "entropy": 4.925964641571045,
      "epoch": 2.625292348813899,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004309146101839893,
      "loss": 4.711,
      "mean_token_accuracy": 0.22738239020109177,
      "num_tokens": 57848251.0,
      "step": 31430
    },
    {
      "entropy": 4.944230842590332,
      "epoch": 2.6257099899766123,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004308931452022911,
      "loss": 4.6173,
      "mean_token_accuracy": 0.23345336467027664,
      "num_tokens": 57856832.0,
      "step": 31435
    },
    {
      "entropy": 5.074586057662964,
      "epoch": 2.626127631139325,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004308716774914957,
      "loss": 4.7594,
      "mean_token_accuracy": 0.2231871709227562,
      "num_tokens": 57866678.0,
      "step": 31440
    },
    {
      "entropy": 5.12024416923523,
      "epoch": 2.626545272302038,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043085020705197906,
      "loss": 4.849,
      "mean_token_accuracy": 0.22289677262306212,
      "num_tokens": 57875764.0,
      "step": 31445
    },
    {
      "entropy": 5.018781089782715,
      "epoch": 2.626962913464751,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043082873388411696,
      "loss": 4.7032,
      "mean_token_accuracy": 0.22746661454439163,
      "num_tokens": 57884964.0,
      "step": 31450
    },
    {
      "entropy": 5.079987382888794,
      "epoch": 2.627380554627464,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00043080725798828526,
      "loss": 4.762,
      "mean_token_accuracy": 0.22201869040727615,
      "num_tokens": 57893625.0,
      "step": 31455
    },
    {
      "entropy": 5.013741970062256,
      "epoch": 2.6277981957901773,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004307857793648599,
      "loss": 4.7034,
      "mean_token_accuracy": 0.23324645161628724,
      "num_tokens": 57902267.0,
      "step": 31460
    },
    {
      "entropy": 4.99457745552063,
      "epoch": 2.62821583695289,
      "grad_norm": 1.125,
      "learning_rate": 0.0004307642980142171,
      "loss": 4.5948,
      "mean_token_accuracy": 0.23403268605470656,
      "num_tokens": 57910681.0,
      "step": 31465
    },
    {
      "entropy": 5.053487300872803,
      "epoch": 2.628633478115603,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043074281393673265,
      "loss": 4.7476,
      "mean_token_accuracy": 0.22350814789533616,
      "num_tokens": 57920589.0,
      "step": 31470
    },
    {
      "entropy": 4.963258934020996,
      "epoch": 2.629051119278316,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043072132713278273,
      "loss": 4.674,
      "mean_token_accuracy": 0.22708968818187714,
      "num_tokens": 57930095.0,
      "step": 31475
    },
    {
      "entropy": 4.988991737365723,
      "epoch": 2.629468760441029,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043069983760274367,
      "loss": 4.6974,
      "mean_token_accuracy": 0.2239366129040718,
      "num_tokens": 57939615.0,
      "step": 31480
    },
    {
      "entropy": 5.088808202743531,
      "epoch": 2.629886401603742,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043067834534699144,
      "loss": 4.7104,
      "mean_token_accuracy": 0.22946594655513763,
      "num_tokens": 57948843.0,
      "step": 31485
    },
    {
      "entropy": 5.054540491104126,
      "epoch": 2.630304042766455,
      "grad_norm": 1.234375,
      "learning_rate": 0.00043065685036590244,
      "loss": 4.8193,
      "mean_token_accuracy": 0.2145104631781578,
      "num_tokens": 57958510.0,
      "step": 31490
    },
    {
      "entropy": 4.99382700920105,
      "epoch": 2.630721683929168,
      "grad_norm": 1.046875,
      "learning_rate": 0.00043063535265985283,
      "loss": 4.5345,
      "mean_token_accuracy": 0.23918339908123015,
      "num_tokens": 57968129.0,
      "step": 31495
    },
    {
      "entropy": 5.021771955490112,
      "epoch": 2.631139325091881,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043061385222921915,
      "loss": 4.686,
      "mean_token_accuracy": 0.21624254137277604,
      "num_tokens": 57977920.0,
      "step": 31500
    },
    {
      "entropy": 5.04931492805481,
      "epoch": 2.631556966254594,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004305923490743777,
      "loss": 4.7581,
      "mean_token_accuracy": 0.22112524658441543,
      "num_tokens": 57987216.0,
      "step": 31505
    },
    {
      "entropy": 5.0237414836883545,
      "epoch": 2.631974607417307,
      "grad_norm": 1.125,
      "learning_rate": 0.0004305708431957048,
      "loss": 4.6564,
      "mean_token_accuracy": 0.22761017382144927,
      "num_tokens": 57996558.0,
      "step": 31510
    },
    {
      "entropy": 4.982155656814575,
      "epoch": 2.63239224858002,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004305493345935769,
      "loss": 4.6315,
      "mean_token_accuracy": 0.2306478276848793,
      "num_tokens": 58005220.0,
      "step": 31515
    },
    {
      "entropy": 5.0077070713043215,
      "epoch": 2.632809889742733,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00043052782326837084,
      "loss": 4.5945,
      "mean_token_accuracy": 0.24160914719104767,
      "num_tokens": 58014841.0,
      "step": 31520
    },
    {
      "entropy": 4.956746625900268,
      "epoch": 2.633227530905446,
      "grad_norm": 1.046875,
      "learning_rate": 0.000430506309220463,
      "loss": 4.6674,
      "mean_token_accuracy": 0.23376851826906203,
      "num_tokens": 58023797.0,
      "step": 31525
    },
    {
      "entropy": 4.9735595226287845,
      "epoch": 2.6336451720681593,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004304847924502301,
      "loss": 4.6915,
      "mean_token_accuracy": 0.23198364526033402,
      "num_tokens": 58032931.0,
      "step": 31530
    },
    {
      "entropy": 5.059205150604248,
      "epoch": 2.634062813230872,
      "grad_norm": 1.046875,
      "learning_rate": 0.00043046327295804865,
      "loss": 4.6585,
      "mean_token_accuracy": 0.23556506931781768,
      "num_tokens": 58042171.0,
      "step": 31535
    },
    {
      "entropy": 4.971545791625976,
      "epoch": 2.634480454393585,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004304417507442955,
      "loss": 4.6641,
      "mean_token_accuracy": 0.2322562262415886,
      "num_tokens": 58050610.0,
      "step": 31540
    },
    {
      "entropy": 4.968385219573975,
      "epoch": 2.634898095556298,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004304202258093473,
      "loss": 4.6727,
      "mean_token_accuracy": 0.23367782980203627,
      "num_tokens": 58059962.0,
      "step": 31545
    },
    {
      "entropy": 4.933579206466675,
      "epoch": 2.635315736719011,
      "grad_norm": 1.03125,
      "learning_rate": 0.00043039869815358114,
      "loss": 4.5854,
      "mean_token_accuracy": 0.23485112935304642,
      "num_tokens": 58069265.0,
      "step": 31550
    },
    {
      "entropy": 4.985416555404663,
      "epoch": 2.6357333778817242,
      "grad_norm": 1.125,
      "learning_rate": 0.0004303771677773736,
      "loss": 4.6068,
      "mean_token_accuracy": 0.2323955252766609,
      "num_tokens": 58078931.0,
      "step": 31555
    },
    {
      "entropy": 4.964158248901367,
      "epoch": 2.636151019044437,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004303556346811018,
      "loss": 4.6012,
      "mean_token_accuracy": 0.23479025661945344,
      "num_tokens": 58088110.0,
      "step": 31560
    },
    {
      "entropy": 4.992136907577515,
      "epoch": 2.63656866020715,
      "grad_norm": 1.125,
      "learning_rate": 0.0004303340988651425,
      "loss": 4.6995,
      "mean_token_accuracy": 0.23170594573020936,
      "num_tokens": 58096975.0,
      "step": 31565
    },
    {
      "entropy": 5.02623610496521,
      "epoch": 2.636986301369863,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004303125603298729,
      "loss": 4.7123,
      "mean_token_accuracy": 0.2233460307121277,
      "num_tokens": 58106633.0,
      "step": 31570
    },
    {
      "entropy": 5.113806056976318,
      "epoch": 2.637403942532576,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043029101907567004,
      "loss": 4.7621,
      "mean_token_accuracy": 0.22039113193750381,
      "num_tokens": 58115705.0,
      "step": 31575
    },
    {
      "entropy": 5.05079140663147,
      "epoch": 2.637821583695289,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004302694751029108,
      "loss": 4.6902,
      "mean_token_accuracy": 0.2277979791164398,
      "num_tokens": 58124191.0,
      "step": 31580
    },
    {
      "entropy": 4.993979167938233,
      "epoch": 2.638239224858002,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004302479284119727,
      "loss": 4.7547,
      "mean_token_accuracy": 0.2238094314932823,
      "num_tokens": 58133911.0,
      "step": 31585
    },
    {
      "entropy": 4.97399411201477,
      "epoch": 2.638656866020715,
      "grad_norm": 1.03125,
      "learning_rate": 0.00043022637900323263,
      "loss": 4.7176,
      "mean_token_accuracy": 0.21613825112581253,
      "num_tokens": 58143299.0,
      "step": 31590
    },
    {
      "entropy": 4.997416400909424,
      "epoch": 2.639074507183428,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004302048268770679,
      "loss": 4.6322,
      "mean_token_accuracy": 0.23289915323257446,
      "num_tokens": 58151890.0,
      "step": 31595
    },
    {
      "entropy": 5.009371328353882,
      "epoch": 2.639492148346141,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043018327203385594,
      "loss": 4.6431,
      "mean_token_accuracy": 0.23494143784046173,
      "num_tokens": 58161381.0,
      "step": 31600
    },
    {
      "entropy": 5.0578795909881595,
      "epoch": 2.639909789508854,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000430161714473974,
      "loss": 4.7729,
      "mean_token_accuracy": 0.22012257874011992,
      "num_tokens": 58170818.0,
      "step": 31605
    },
    {
      "entropy": 5.005090999603271,
      "epoch": 2.640327430671567,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043014015419779947,
      "loss": 4.7011,
      "mean_token_accuracy": 0.23502954542636872,
      "num_tokens": 58180050.0,
      "step": 31610
    },
    {
      "entropy": 5.004791259765625,
      "epoch": 2.6407450718342798,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004301185912057097,
      "loss": 4.6402,
      "mean_token_accuracy": 0.23426138311624528,
      "num_tokens": 58188252.0,
      "step": 31615
    },
    {
      "entropy": 5.043140411376953,
      "epoch": 2.641162712996993,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00043009702549808243,
      "loss": 4.702,
      "mean_token_accuracy": 0.2228125363588333,
      "num_tokens": 58196772.0,
      "step": 31620
    },
    {
      "entropy": 5.020703554153442,
      "epoch": 2.641580354159706,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043007545707529496,
      "loss": 4.7823,
      "mean_token_accuracy": 0.22304383665323257,
      "num_tokens": 58205986.0,
      "step": 31625
    },
    {
      "entropy": 5.110866069793701,
      "epoch": 2.641997995322419,
      "grad_norm": 1.03125,
      "learning_rate": 0.000430053885937725,
      "loss": 4.7406,
      "mean_token_accuracy": 0.22311634719371795,
      "num_tokens": 58214596.0,
      "step": 31630
    },
    {
      "entropy": 5.064557647705078,
      "epoch": 2.642415636485132,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043003231208575004,
      "loss": 4.7332,
      "mean_token_accuracy": 0.22409875839948654,
      "num_tokens": 58224011.0,
      "step": 31635
    },
    {
      "entropy": 4.959453582763672,
      "epoch": 2.6428332776478447,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043001073551974796,
      "loss": 4.6658,
      "mean_token_accuracy": 0.23484339863061904,
      "num_tokens": 58233175.0,
      "step": 31640
    },
    {
      "entropy": 5.098942518234253,
      "epoch": 2.643250918810558,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042998915624009625,
      "loss": 4.767,
      "mean_token_accuracy": 0.22712542116641998,
      "num_tokens": 58242496.0,
      "step": 31645
    },
    {
      "entropy": 5.055347919464111,
      "epoch": 2.643668559973271,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004299675742471729,
      "loss": 4.6577,
      "mean_token_accuracy": 0.22786849737167358,
      "num_tokens": 58251361.0,
      "step": 31650
    },
    {
      "entropy": 5.062995195388794,
      "epoch": 2.644086201135984,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042994598954135553,
      "loss": 4.7395,
      "mean_token_accuracy": 0.22263948917388915,
      "num_tokens": 58260892.0,
      "step": 31655
    },
    {
      "entropy": 5.0598962783813475,
      "epoch": 2.644503842298697,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004299244021230222,
      "loss": 4.6833,
      "mean_token_accuracy": 0.22772133499383926,
      "num_tokens": 58270109.0,
      "step": 31660
    },
    {
      "entropy": 5.096974182128906,
      "epoch": 2.64492148346141,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004299028119925507,
      "loss": 4.8042,
      "mean_token_accuracy": 0.22788797020912172,
      "num_tokens": 58278700.0,
      "step": 31665
    },
    {
      "entropy": 5.0716033458709715,
      "epoch": 2.645339124624123,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042988121915031894,
      "loss": 4.7268,
      "mean_token_accuracy": 0.23150526881217956,
      "num_tokens": 58288671.0,
      "step": 31670
    },
    {
      "entropy": 5.035849094390869,
      "epoch": 2.645756765786836,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004298596235967051,
      "loss": 4.7267,
      "mean_token_accuracy": 0.2307454913854599,
      "num_tokens": 58297428.0,
      "step": 31675
    },
    {
      "entropy": 5.081360101699829,
      "epoch": 2.646174406949549,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004298380253320871,
      "loss": 4.7725,
      "mean_token_accuracy": 0.22549076825380326,
      "num_tokens": 58306863.0,
      "step": 31680
    },
    {
      "entropy": 5.016495275497436,
      "epoch": 2.646592048112262,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004298164243568431,
      "loss": 4.7424,
      "mean_token_accuracy": 0.22765709161758424,
      "num_tokens": 58315701.0,
      "step": 31685
    },
    {
      "entropy": 5.022803163528442,
      "epoch": 2.647009689274975,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004297948206713513,
      "loss": 4.7999,
      "mean_token_accuracy": 0.22282288670539857,
      "num_tokens": 58324805.0,
      "step": 31690
    },
    {
      "entropy": 5.096542501449585,
      "epoch": 2.647427330437688,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042977321427598974,
      "loss": 4.6895,
      "mean_token_accuracy": 0.22139762043952943,
      "num_tokens": 58332885.0,
      "step": 31695
    },
    {
      "entropy": 5.0085564136505125,
      "epoch": 2.647844971600401,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004297516051711368,
      "loss": 4.7231,
      "mean_token_accuracy": 0.21987627744674682,
      "num_tokens": 58343447.0,
      "step": 31700
    },
    {
      "entropy": 5.086364364624023,
      "epoch": 2.648262612763114,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004297299933571708,
      "loss": 4.7511,
      "mean_token_accuracy": 0.2202850729227066,
      "num_tokens": 58352807.0,
      "step": 31705
    },
    {
      "entropy": 5.054631662368775,
      "epoch": 2.6486802539258267,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004297083788344699,
      "loss": 4.734,
      "mean_token_accuracy": 0.22458574026823044,
      "num_tokens": 58361683.0,
      "step": 31710
    },
    {
      "entropy": 5.106881189346313,
      "epoch": 2.64909789508854,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042968676160341275,
      "loss": 4.7529,
      "mean_token_accuracy": 0.22705814987421036,
      "num_tokens": 58370760.0,
      "step": 31715
    },
    {
      "entropy": 4.985549402236939,
      "epoch": 2.649515536251253,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004296651416643775,
      "loss": 4.6519,
      "mean_token_accuracy": 0.2324380174279213,
      "num_tokens": 58379931.0,
      "step": 31720
    },
    {
      "entropy": 4.947374677658081,
      "epoch": 2.649933177413966,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042964351901774285,
      "loss": 4.5649,
      "mean_token_accuracy": 0.24029105454683303,
      "num_tokens": 58388320.0,
      "step": 31725
    },
    {
      "entropy": 4.973574113845825,
      "epoch": 2.650350818576679,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004296218936638873,
      "loss": 4.7138,
      "mean_token_accuracy": 0.22193968594074248,
      "num_tokens": 58397610.0,
      "step": 31730
    },
    {
      "entropy": 5.056680393218994,
      "epoch": 2.6507684597393917,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042960026560318927,
      "loss": 4.6827,
      "mean_token_accuracy": 0.2274975448846817,
      "num_tokens": 58407156.0,
      "step": 31735
    },
    {
      "entropy": 5.16344199180603,
      "epoch": 2.651186100902105,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042957863483602756,
      "loss": 4.9202,
      "mean_token_accuracy": 0.21084395796060562,
      "num_tokens": 58416900.0,
      "step": 31740
    },
    {
      "entropy": 5.0194603443145756,
      "epoch": 2.6516037420648177,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042955700136278077,
      "loss": 4.5968,
      "mean_token_accuracy": 0.22805212289094925,
      "num_tokens": 58425714.0,
      "step": 31745
    },
    {
      "entropy": 4.911818027496338,
      "epoch": 2.652021383227531,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004295353651838276,
      "loss": 4.5341,
      "mean_token_accuracy": 0.23720824271440505,
      "num_tokens": 58433910.0,
      "step": 31750
    },
    {
      "entropy": 5.031021070480347,
      "epoch": 2.652439024390244,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042951372629954695,
      "loss": 4.7986,
      "mean_token_accuracy": 0.22039948403835297,
      "num_tokens": 58443339.0,
      "step": 31755
    },
    {
      "entropy": 5.099263525009155,
      "epoch": 2.6528566655529566,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004294920847103174,
      "loss": 4.6791,
      "mean_token_accuracy": 0.23444096148014068,
      "num_tokens": 58451481.0,
      "step": 31760
    },
    {
      "entropy": 5.010876083374024,
      "epoch": 2.65327430671567,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004294704404165181,
      "loss": 4.7091,
      "mean_token_accuracy": 0.22587839812040328,
      "num_tokens": 58461970.0,
      "step": 31765
    },
    {
      "entropy": 5.039813041687012,
      "epoch": 2.653691947878383,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042944879341852763,
      "loss": 4.7321,
      "mean_token_accuracy": 0.22233611792325975,
      "num_tokens": 58471210.0,
      "step": 31770
    },
    {
      "entropy": 5.008195114135742,
      "epoch": 2.654109589041096,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004294271437167252,
      "loss": 4.6295,
      "mean_token_accuracy": 0.23136118650436402,
      "num_tokens": 58480808.0,
      "step": 31775
    },
    {
      "entropy": 5.051388549804687,
      "epoch": 2.6545272302038088,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00042940549131148976,
      "loss": 4.8483,
      "mean_token_accuracy": 0.2132920578122139,
      "num_tokens": 58490514.0,
      "step": 31780
    },
    {
      "entropy": 5.130674600601196,
      "epoch": 2.654944871366522,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004293838362032003,
      "loss": 4.8128,
      "mean_token_accuracy": 0.21210913807153703,
      "num_tokens": 58499894.0,
      "step": 31785
    },
    {
      "entropy": 5.141251420974731,
      "epoch": 2.655362512529235,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042936217839223596,
      "loss": 4.7724,
      "mean_token_accuracy": 0.22770626842975616,
      "num_tokens": 58510531.0,
      "step": 31790
    },
    {
      "entropy": 4.956698560714722,
      "epoch": 2.655780153691948,
      "grad_norm": 1.21875,
      "learning_rate": 0.00042934051787897593,
      "loss": 4.6424,
      "mean_token_accuracy": 0.22781771123409272,
      "num_tokens": 58519542.0,
      "step": 31795
    },
    {
      "entropy": 4.953580665588379,
      "epoch": 2.656197794854661,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00042931885466379935,
      "loss": 4.6223,
      "mean_token_accuracy": 0.23424438536167144,
      "num_tokens": 58528883.0,
      "step": 31800
    },
    {
      "entropy": 4.992829084396362,
      "epoch": 2.6566154360173737,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004292971887470854,
      "loss": 4.6895,
      "mean_token_accuracy": 0.23615407943725586,
      "num_tokens": 58538185.0,
      "step": 31805
    },
    {
      "entropy": 5.056937265396118,
      "epoch": 2.657033077180087,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042927552012921353,
      "loss": 4.7739,
      "mean_token_accuracy": 0.2287200540304184,
      "num_tokens": 58547933.0,
      "step": 31810
    },
    {
      "entropy": 5.028302526473999,
      "epoch": 2.6574507183428,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042925384881056284,
      "loss": 4.595,
      "mean_token_accuracy": 0.23739609122276306,
      "num_tokens": 58557291.0,
      "step": 31815
    },
    {
      "entropy": 4.937295436859131,
      "epoch": 2.657868359505513,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042923217479151296,
      "loss": 4.5873,
      "mean_token_accuracy": 0.238790924847126,
      "num_tokens": 58565693.0,
      "step": 31820
    },
    {
      "entropy": 5.005168533325195,
      "epoch": 2.658286000668226,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042921049807244317,
      "loss": 4.6843,
      "mean_token_accuracy": 0.23014656454324722,
      "num_tokens": 58574557.0,
      "step": 31825
    },
    {
      "entropy": 5.086208868026733,
      "epoch": 2.6587036418309387,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000429188818653733,
      "loss": 4.7589,
      "mean_token_accuracy": 0.22666535377502442,
      "num_tokens": 58582820.0,
      "step": 31830
    },
    {
      "entropy": 4.94911298751831,
      "epoch": 2.659121282993652,
      "grad_norm": 1.25,
      "learning_rate": 0.0004291671365357619,
      "loss": 4.6079,
      "mean_token_accuracy": 0.22396999150514602,
      "num_tokens": 58591559.0,
      "step": 31835
    },
    {
      "entropy": 5.095627355575561,
      "epoch": 2.6595389241563647,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042914545171890964,
      "loss": 4.7516,
      "mean_token_accuracy": 0.2220725730061531,
      "num_tokens": 58600788.0,
      "step": 31840
    },
    {
      "entropy": 4.994692373275757,
      "epoch": 2.659956565319078,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042912376420355553,
      "loss": 4.5616,
      "mean_token_accuracy": 0.2421793669462204,
      "num_tokens": 58609742.0,
      "step": 31845
    },
    {
      "entropy": 5.006908750534057,
      "epoch": 2.660374206481791,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004291020739900796,
      "loss": 4.6843,
      "mean_token_accuracy": 0.2252035692334175,
      "num_tokens": 58619162.0,
      "step": 31850
    },
    {
      "entropy": 4.98219108581543,
      "epoch": 2.6607918476445036,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004290803810788612,
      "loss": 4.7143,
      "mean_token_accuracy": 0.22337122559547423,
      "num_tokens": 58628873.0,
      "step": 31855
    },
    {
      "entropy": 5.017495203018188,
      "epoch": 2.661209488807217,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004290586854702804,
      "loss": 4.7055,
      "mean_token_accuracy": 0.22789791077375413,
      "num_tokens": 58638180.0,
      "step": 31860
    },
    {
      "entropy": 4.967859268188477,
      "epoch": 2.6616271299699297,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042903698716471677,
      "loss": 4.6782,
      "mean_token_accuracy": 0.22798793762922287,
      "num_tokens": 58647303.0,
      "step": 31865
    },
    {
      "entropy": 5.156368923187256,
      "epoch": 2.662044771132643,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004290152861625503,
      "loss": 4.7953,
      "mean_token_accuracy": 0.21922989338636398,
      "num_tokens": 58656724.0,
      "step": 31870
    },
    {
      "entropy": 5.111675262451172,
      "epoch": 2.6624624122953557,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042899358246416085,
      "loss": 4.8108,
      "mean_token_accuracy": 0.22211849838495254,
      "num_tokens": 58666670.0,
      "step": 31875
    },
    {
      "entropy": 5.0000073432922365,
      "epoch": 2.6628800534580686,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042897187606992846,
      "loss": 4.6287,
      "mean_token_accuracy": 0.2292042687535286,
      "num_tokens": 58675608.0,
      "step": 31880
    },
    {
      "entropy": 5.014566659927368,
      "epoch": 2.663297694620782,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004289501669802329,
      "loss": 4.7581,
      "mean_token_accuracy": 0.22235042750835418,
      "num_tokens": 58685470.0,
      "step": 31885
    },
    {
      "entropy": 5.136612462997436,
      "epoch": 2.663715335783495,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004289284551954545,
      "loss": 4.807,
      "mean_token_accuracy": 0.2195862278342247,
      "num_tokens": 58695610.0,
      "step": 31890
    },
    {
      "entropy": 5.069318675994873,
      "epoch": 2.664132976946208,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004289067407159731,
      "loss": 4.7409,
      "mean_token_accuracy": 0.22656115293502807,
      "num_tokens": 58705608.0,
      "step": 31895
    },
    {
      "entropy": 4.979587364196777,
      "epoch": 2.6645506181089207,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042888502354216897,
      "loss": 4.6298,
      "mean_token_accuracy": 0.2336413338780403,
      "num_tokens": 58714575.0,
      "step": 31900
    },
    {
      "entropy": 5.054866170883178,
      "epoch": 2.664968259271634,
      "grad_norm": 1.046875,
      "learning_rate": 0.00042886330367442234,
      "loss": 4.8068,
      "mean_token_accuracy": 0.21917077898979187,
      "num_tokens": 58724748.0,
      "step": 31905
    },
    {
      "entropy": 5.107120561599731,
      "epoch": 2.6653859004343468,
      "grad_norm": 1.046875,
      "learning_rate": 0.00042884158111311326,
      "loss": 4.8134,
      "mean_token_accuracy": 0.21857843697071075,
      "num_tokens": 58734388.0,
      "step": 31910
    },
    {
      "entropy": 5.0552685260772705,
      "epoch": 2.66580354159706,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004288198558586222,
      "loss": 4.6319,
      "mean_token_accuracy": 0.23643114864826204,
      "num_tokens": 58742328.0,
      "step": 31915
    },
    {
      "entropy": 5.01777024269104,
      "epoch": 2.666221182759773,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004287981279113294,
      "loss": 4.658,
      "mean_token_accuracy": 0.22715459018945694,
      "num_tokens": 58750788.0,
      "step": 31920
    },
    {
      "entropy": 4.959821891784668,
      "epoch": 2.6666388239224856,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004287763972716152,
      "loss": 4.6702,
      "mean_token_accuracy": 0.23270942866802216,
      "num_tokens": 58759305.0,
      "step": 31925
    },
    {
      "entropy": 5.070625257492066,
      "epoch": 2.667056465085199,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042875466393986017,
      "loss": 4.699,
      "mean_token_accuracy": 0.224679197371006,
      "num_tokens": 58768358.0,
      "step": 31930
    },
    {
      "entropy": 4.9952727317810055,
      "epoch": 2.6674741062479117,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004287329279164445,
      "loss": 4.6696,
      "mean_token_accuracy": 0.23478995710611344,
      "num_tokens": 58778005.0,
      "step": 31935
    },
    {
      "entropy": 5.033133316040039,
      "epoch": 2.667891747410625,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00042871118920174907,
      "loss": 4.833,
      "mean_token_accuracy": 0.21035777777433395,
      "num_tokens": 58787457.0,
      "step": 31940
    },
    {
      "entropy": 5.0382716178894045,
      "epoch": 2.6683093885733378,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004286894477961541,
      "loss": 4.6909,
      "mean_token_accuracy": 0.22739552706480026,
      "num_tokens": 58796671.0,
      "step": 31945
    },
    {
      "entropy": 4.998365497589111,
      "epoch": 2.6687270297360506,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042866770370004044,
      "loss": 4.67,
      "mean_token_accuracy": 0.23368497639894487,
      "num_tokens": 58805685.0,
      "step": 31950
    },
    {
      "entropy": 5.106556797027588,
      "epoch": 2.669144670898764,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004286459569137886,
      "loss": 4.8177,
      "mean_token_accuracy": 0.21502392143011093,
      "num_tokens": 58816453.0,
      "step": 31955
    },
    {
      "entropy": 5.011555624008179,
      "epoch": 2.6695623120614766,
      "grad_norm": 1.0,
      "learning_rate": 0.0004286242074377794,
      "loss": 4.5772,
      "mean_token_accuracy": 0.23946287781000136,
      "num_tokens": 58825684.0,
      "step": 31960
    },
    {
      "entropy": 4.937980461120605,
      "epoch": 2.66997995322419,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004286024552723935,
      "loss": 4.6617,
      "mean_token_accuracy": 0.23007443696260452,
      "num_tokens": 58834481.0,
      "step": 31965
    },
    {
      "entropy": 4.9251645565032955,
      "epoch": 2.6703975943869027,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042858070041801175,
      "loss": 4.6108,
      "mean_token_accuracy": 0.23815826773643495,
      "num_tokens": 58843631.0,
      "step": 31970
    },
    {
      "entropy": 5.063185739517212,
      "epoch": 2.6708152355496155,
      "grad_norm": 1.125,
      "learning_rate": 0.00042855894287501493,
      "loss": 4.6749,
      "mean_token_accuracy": 0.22552505880594254,
      "num_tokens": 58852415.0,
      "step": 31975
    },
    {
      "entropy": 5.023036670684815,
      "epoch": 2.671232876712329,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00042853718264378404,
      "loss": 4.6542,
      "mean_token_accuracy": 0.2333907291293144,
      "num_tokens": 58862016.0,
      "step": 31980
    },
    {
      "entropy": 4.9609071731567385,
      "epoch": 2.671650517875042,
      "grad_norm": 1.125,
      "learning_rate": 0.00042851541972469996,
      "loss": 4.6359,
      "mean_token_accuracy": 0.23218447268009185,
      "num_tokens": 58871163.0,
      "step": 31985
    },
    {
      "entropy": 5.01223087310791,
      "epoch": 2.672068159037755,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042849365411814355,
      "loss": 4.6722,
      "mean_token_accuracy": 0.23047930300235747,
      "num_tokens": 58880075.0,
      "step": 31990
    },
    {
      "entropy": 4.990357732772827,
      "epoch": 2.6724858002004677,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004284718858244961,
      "loss": 4.6336,
      "mean_token_accuracy": 0.24252256006002426,
      "num_tokens": 58887835.0,
      "step": 31995
    },
    {
      "entropy": 4.973804140090943,
      "epoch": 2.672903441363181,
      "grad_norm": 1.046875,
      "learning_rate": 0.00042845011484413846,
      "loss": 4.6749,
      "mean_token_accuracy": 0.2249221310019493,
      "num_tokens": 58897085.0,
      "step": 32000
    },
    {
      "entropy": 4.899964809417725,
      "epoch": 2.6733210825258937,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004284283411774519,
      "loss": 4.6437,
      "mean_token_accuracy": 0.2378702402114868,
      "num_tokens": 58905768.0,
      "step": 32005
    },
    {
      "entropy": 5.13639006614685,
      "epoch": 2.673738723688607,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00042840656482481755,
      "loss": 4.8063,
      "mean_token_accuracy": 0.22274285256862641,
      "num_tokens": 58915773.0,
      "step": 32010
    },
    {
      "entropy": 5.1959514141082765,
      "epoch": 2.67415636485132,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004283847857866167,
      "loss": 4.7676,
      "mean_token_accuracy": 0.22249847054481506,
      "num_tokens": 58925586.0,
      "step": 32015
    },
    {
      "entropy": 5.036986446380615,
      "epoch": 2.6745740060140326,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042836300406323035,
      "loss": 4.7242,
      "mean_token_accuracy": 0.21935490220785142,
      "num_tokens": 58934777.0,
      "step": 32020
    },
    {
      "entropy": 4.951601552963257,
      "epoch": 2.674991647176746,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042834121965504014,
      "loss": 4.6995,
      "mean_token_accuracy": 0.22794897109270096,
      "num_tokens": 58943825.0,
      "step": 32025
    },
    {
      "entropy": 5.073218917846679,
      "epoch": 2.6754092883394587,
      "grad_norm": 1.046875,
      "learning_rate": 0.00042831943256242726,
      "loss": 4.6756,
      "mean_token_accuracy": 0.23443885445594786,
      "num_tokens": 58952880.0,
      "step": 32030
    },
    {
      "entropy": 5.046646595001221,
      "epoch": 2.675826929502172,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004282976427857732,
      "loss": 4.6489,
      "mean_token_accuracy": 0.22766717672348022,
      "num_tokens": 58962499.0,
      "step": 32035
    },
    {
      "entropy": 4.983708381652832,
      "epoch": 2.6762445706648847,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004282758503254594,
      "loss": 4.6549,
      "mean_token_accuracy": 0.23399094939231874,
      "num_tokens": 58972184.0,
      "step": 32040
    },
    {
      "entropy": 4.990020322799682,
      "epoch": 2.6766622118275976,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042825405518186725,
      "loss": 4.5726,
      "mean_token_accuracy": 0.23653278052806853,
      "num_tokens": 58980638.0,
      "step": 32045
    },
    {
      "entropy": 5.0088782787323,
      "epoch": 2.677079852990311,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042823225735537845,
      "loss": 4.7554,
      "mean_token_accuracy": 0.21762894839048386,
      "num_tokens": 58989821.0,
      "step": 32050
    },
    {
      "entropy": 5.003710842132568,
      "epoch": 2.6774974941530236,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004282104568463745,
      "loss": 4.7244,
      "mean_token_accuracy": 0.22545998096466063,
      "num_tokens": 58998444.0,
      "step": 32055
    },
    {
      "entropy": 5.020945835113525,
      "epoch": 2.677915135315737,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004281886536552372,
      "loss": 4.7406,
      "mean_token_accuracy": 0.23074116557836533,
      "num_tokens": 59007870.0,
      "step": 32060
    },
    {
      "entropy": 5.097191572189331,
      "epoch": 2.6783327764784497,
      "grad_norm": 1.078125,
      "learning_rate": 0.000428166847782348,
      "loss": 4.6842,
      "mean_token_accuracy": 0.2282635062932968,
      "num_tokens": 59017443.0,
      "step": 32065
    },
    {
      "entropy": 5.019794654846192,
      "epoch": 2.6787504176411625,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004281450392280889,
      "loss": 4.7737,
      "mean_token_accuracy": 0.22339832186698913,
      "num_tokens": 59027256.0,
      "step": 32070
    },
    {
      "entropy": 5.0647135257720945,
      "epoch": 2.6791680588038758,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004281232279928414,
      "loss": 4.7185,
      "mean_token_accuracy": 0.2243145763874054,
      "num_tokens": 59036520.0,
      "step": 32075
    },
    {
      "entropy": 5.0949078559875485,
      "epoch": 2.6795856999665886,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004281014140769875,
      "loss": 4.7705,
      "mean_token_accuracy": 0.2229922190308571,
      "num_tokens": 59045823.0,
      "step": 32080
    },
    {
      "entropy": 5.053846740722657,
      "epoch": 2.680003341129302,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042807959748090914,
      "loss": 4.6976,
      "mean_token_accuracy": 0.23102566450834275,
      "num_tokens": 59055316.0,
      "step": 32085
    },
    {
      "entropy": 4.904661464691162,
      "epoch": 2.6804209822920146,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004280577782049881,
      "loss": 4.6134,
      "mean_token_accuracy": 0.23162386268377305,
      "num_tokens": 59064423.0,
      "step": 32090
    },
    {
      "entropy": 5.0630255222320555,
      "epoch": 2.6808386234547275,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042803595624960647,
      "loss": 4.8302,
      "mean_token_accuracy": 0.21091433614492416,
      "num_tokens": 59074212.0,
      "step": 32095
    },
    {
      "entropy": 5.036915922164917,
      "epoch": 2.6812562646174407,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042801413161514613,
      "loss": 4.6512,
      "mean_token_accuracy": 0.2407069906592369,
      "num_tokens": 59083098.0,
      "step": 32100
    },
    {
      "entropy": 5.008153200149536,
      "epoch": 2.681673905780154,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042799230430198935,
      "loss": 4.6958,
      "mean_token_accuracy": 0.2339257761836052,
      "num_tokens": 59091709.0,
      "step": 32105
    },
    {
      "entropy": 5.076227760314941,
      "epoch": 2.6820915469428668,
      "grad_norm": 1.234375,
      "learning_rate": 0.000427970474310518,
      "loss": 4.7327,
      "mean_token_accuracy": 0.227737395465374,
      "num_tokens": 59100495.0,
      "step": 32110
    },
    {
      "entropy": 5.086587285995483,
      "epoch": 2.6825091881055796,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00042794864164111447,
      "loss": 4.7753,
      "mean_token_accuracy": 0.22447052001953124,
      "num_tokens": 59109507.0,
      "step": 32115
    },
    {
      "entropy": 5.120188808441162,
      "epoch": 2.682926829268293,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004279268062941608,
      "loss": 4.6891,
      "mean_token_accuracy": 0.23598723113536835,
      "num_tokens": 59119132.0,
      "step": 32120
    },
    {
      "entropy": 5.064363098144531,
      "epoch": 2.6833444704310057,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004279049682700394,
      "loss": 4.745,
      "mean_token_accuracy": 0.22511315941810608,
      "num_tokens": 59128049.0,
      "step": 32125
    },
    {
      "entropy": 5.084659242630005,
      "epoch": 2.683762111593719,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042788312756913234,
      "loss": 4.6862,
      "mean_token_accuracy": 0.22091220319271088,
      "num_tokens": 59137292.0,
      "step": 32130
    },
    {
      "entropy": 5.077216148376465,
      "epoch": 2.6841797527564317,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004278612841918222,
      "loss": 4.7721,
      "mean_token_accuracy": 0.22703766971826553,
      "num_tokens": 59146210.0,
      "step": 32135
    },
    {
      "entropy": 5.032685565948486,
      "epoch": 2.6845973939191445,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004278394381384912,
      "loss": 4.6753,
      "mean_token_accuracy": 0.23070713132619858,
      "num_tokens": 59155968.0,
      "step": 32140
    },
    {
      "entropy": 5.089319086074829,
      "epoch": 2.685015035081858,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042781758940952194,
      "loss": 4.6996,
      "mean_token_accuracy": 0.22520127147436142,
      "num_tokens": 59165824.0,
      "step": 32145
    },
    {
      "entropy": 5.044409990310669,
      "epoch": 2.6854326762445706,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004277957380052968,
      "loss": 4.6806,
      "mean_token_accuracy": 0.23699649423360825,
      "num_tokens": 59174467.0,
      "step": 32150
    },
    {
      "entropy": 4.961599636077881,
      "epoch": 2.685850317407284,
      "grad_norm": 1.046875,
      "learning_rate": 0.00042777388392619835,
      "loss": 4.6008,
      "mean_token_accuracy": 0.23043678551912308,
      "num_tokens": 59184159.0,
      "step": 32155
    },
    {
      "entropy": 5.031974744796753,
      "epoch": 2.6862679585699967,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004277520271726092,
      "loss": 4.7032,
      "mean_token_accuracy": 0.2274893820285797,
      "num_tokens": 59193488.0,
      "step": 32160
    },
    {
      "entropy": 5.053756475448608,
      "epoch": 2.6866855997327095,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004277301677449119,
      "loss": 4.715,
      "mean_token_accuracy": 0.2313856303691864,
      "num_tokens": 59203643.0,
      "step": 32165
    },
    {
      "entropy": 4.985066509246826,
      "epoch": 2.6871032408954227,
      "grad_norm": 1.125,
      "learning_rate": 0.00042770830564348915,
      "loss": 4.5996,
      "mean_token_accuracy": 0.234512759745121,
      "num_tokens": 59212962.0,
      "step": 32170
    },
    {
      "entropy": 5.0515186309814455,
      "epoch": 2.6875208820581356,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042768644086872384,
      "loss": 4.777,
      "mean_token_accuracy": 0.21736137568950653,
      "num_tokens": 59222798.0,
      "step": 32175
    },
    {
      "entropy": 5.1108255863189695,
      "epoch": 2.687938523220849,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004276645734209984,
      "loss": 4.7994,
      "mean_token_accuracy": 0.22139362543821334,
      "num_tokens": 59232301.0,
      "step": 32180
    },
    {
      "entropy": 5.048775339126587,
      "epoch": 2.6883561643835616,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000427642703300696,
      "loss": 4.6724,
      "mean_token_accuracy": 0.2279712289571762,
      "num_tokens": 59241953.0,
      "step": 32185
    },
    {
      "entropy": 5.104407739639282,
      "epoch": 2.6887738055462744,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004276208305081993,
      "loss": 4.7864,
      "mean_token_accuracy": 0.22008519172668456,
      "num_tokens": 59251297.0,
      "step": 32190
    },
    {
      "entropy": 5.082111024856568,
      "epoch": 2.6891914467089877,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004275989550438912,
      "loss": 4.7644,
      "mean_token_accuracy": 0.22775738537311555,
      "num_tokens": 59260037.0,
      "step": 32195
    },
    {
      "entropy": 5.072155714035034,
      "epoch": 2.6896090878717005,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004275770769081547,
      "loss": 4.79,
      "mean_token_accuracy": 0.22187268137931823,
      "num_tokens": 59269136.0,
      "step": 32200
    },
    {
      "entropy": 5.0187718868255615,
      "epoch": 2.6900267290344138,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004275551961013729,
      "loss": 4.6941,
      "mean_token_accuracy": 0.2267044872045517,
      "num_tokens": 59278211.0,
      "step": 32205
    },
    {
      "entropy": 5.078966999053955,
      "epoch": 2.6904443701971266,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042753331262392866,
      "loss": 4.7364,
      "mean_token_accuracy": 0.21931222826242447,
      "num_tokens": 59286596.0,
      "step": 32210
    },
    {
      "entropy": 5.059680461883545,
      "epoch": 2.6908620113598394,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042751142647620515,
      "loss": 4.6335,
      "mean_token_accuracy": 0.22387561202049255,
      "num_tokens": 59294893.0,
      "step": 32215
    },
    {
      "entropy": 5.01606822013855,
      "epoch": 2.6912796525225526,
      "grad_norm": 1.03125,
      "learning_rate": 0.00042748953765858565,
      "loss": 4.6667,
      "mean_token_accuracy": 0.23255551010370254,
      "num_tokens": 59304621.0,
      "step": 32220
    },
    {
      "entropy": 5.026334810256958,
      "epoch": 2.691697293685266,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004274676461714532,
      "loss": 4.7636,
      "mean_token_accuracy": 0.21545057743787766,
      "num_tokens": 59313429.0,
      "step": 32225
    },
    {
      "entropy": 5.054666471481323,
      "epoch": 2.6921149348479787,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042744575201519115,
      "loss": 4.7514,
      "mean_token_accuracy": 0.2160709023475647,
      "num_tokens": 59323323.0,
      "step": 32230
    },
    {
      "entropy": 5.108335065841675,
      "epoch": 2.6925325760106915,
      "grad_norm": 1.046875,
      "learning_rate": 0.00042742385519018263,
      "loss": 4.7744,
      "mean_token_accuracy": 0.2249945044517517,
      "num_tokens": 59333070.0,
      "step": 32235
    },
    {
      "entropy": 5.082858514785767,
      "epoch": 2.6929502171734048,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004274019556968111,
      "loss": 4.7344,
      "mean_token_accuracy": 0.2266654536128044,
      "num_tokens": 59342303.0,
      "step": 32240
    },
    {
      "entropy": 4.978161239624024,
      "epoch": 2.6933678583361176,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042738005353545986,
      "loss": 4.6714,
      "mean_token_accuracy": 0.2307526007294655,
      "num_tokens": 59351465.0,
      "step": 32245
    },
    {
      "entropy": 4.987631034851074,
      "epoch": 2.693785499498831,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042735814870651235,
      "loss": 4.6831,
      "mean_token_accuracy": 0.22310216277837752,
      "num_tokens": 59360696.0,
      "step": 32250
    },
    {
      "entropy": 5.061803388595581,
      "epoch": 2.6942031406615436,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004273362412103521,
      "loss": 4.7049,
      "mean_token_accuracy": 0.2310768947005272,
      "num_tokens": 59369819.0,
      "step": 32255
    },
    {
      "entropy": 5.054657077789306,
      "epoch": 2.6946207818242565,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004273143310473625,
      "loss": 4.7325,
      "mean_token_accuracy": 0.2250918462872505,
      "num_tokens": 59378599.0,
      "step": 32260
    },
    {
      "entropy": 5.0723443031311035,
      "epoch": 2.6950384229869697,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00042729241821792726,
      "loss": 4.7678,
      "mean_token_accuracy": 0.21890788227319719,
      "num_tokens": 59389598.0,
      "step": 32265
    },
    {
      "entropy": 5.087965106964111,
      "epoch": 2.6954560641496825,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004272705027224299,
      "loss": 4.7995,
      "mean_token_accuracy": 0.223590624332428,
      "num_tokens": 59399279.0,
      "step": 32270
    },
    {
      "entropy": 5.036603546142578,
      "epoch": 2.695873705312396,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004272485845612541,
      "loss": 4.7003,
      "mean_token_accuracy": 0.22945604473352432,
      "num_tokens": 59407601.0,
      "step": 32275
    },
    {
      "entropy": 5.08830075263977,
      "epoch": 2.6962913464751086,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00042722666373478356,
      "loss": 4.7231,
      "mean_token_accuracy": 0.22207076251506805,
      "num_tokens": 59417177.0,
      "step": 32280
    },
    {
      "entropy": 5.005128717422485,
      "epoch": 2.6967089876378214,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042720474024340204,
      "loss": 4.667,
      "mean_token_accuracy": 0.23535743802785875,
      "num_tokens": 59425370.0,
      "step": 32285
    },
    {
      "entropy": 5.023765420913696,
      "epoch": 2.6971266288005347,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004271828140874934,
      "loss": 4.6865,
      "mean_token_accuracy": 0.23045874536037445,
      "num_tokens": 59435272.0,
      "step": 32290
    },
    {
      "entropy": 5.032657289505005,
      "epoch": 2.6975442699632475,
      "grad_norm": 1.046875,
      "learning_rate": 0.00042716088526744127,
      "loss": 4.6755,
      "mean_token_accuracy": 0.23784591555595397,
      "num_tokens": 59443962.0,
      "step": 32295
    },
    {
      "entropy": 4.967190933227539,
      "epoch": 2.6979619111259607,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004271389537836299,
      "loss": 4.6074,
      "mean_token_accuracy": 0.2353512704372406,
      "num_tokens": 59452758.0,
      "step": 32300
    },
    {
      "entropy": 5.037147760391235,
      "epoch": 2.6983795522886735,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042711701963644283,
      "loss": 4.7042,
      "mean_token_accuracy": 0.22273064404726028,
      "num_tokens": 59461280.0,
      "step": 32305
    },
    {
      "entropy": 4.965669775009156,
      "epoch": 2.6987971934513864,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042709508282626424,
      "loss": 4.6623,
      "mean_token_accuracy": 0.22843580991029738,
      "num_tokens": 59470231.0,
      "step": 32310
    },
    {
      "entropy": 5.070265293121338,
      "epoch": 2.6992148346140996,
      "grad_norm": 1.125,
      "learning_rate": 0.00042707314335347827,
      "loss": 4.7992,
      "mean_token_accuracy": 0.221595461666584,
      "num_tokens": 59480788.0,
      "step": 32315
    },
    {
      "entropy": 5.085563278198242,
      "epoch": 2.6996324757768124,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042705120121846877,
      "loss": 4.778,
      "mean_token_accuracy": 0.2194800302386284,
      "num_tokens": 59490920.0,
      "step": 32320
    },
    {
      "entropy": 5.020417881011963,
      "epoch": 2.7000501169395257,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042702925642162,
      "loss": 4.6385,
      "mean_token_accuracy": 0.23260127156972885,
      "num_tokens": 59499324.0,
      "step": 32325
    },
    {
      "entropy": 5.080379343032837,
      "epoch": 2.7004677581022385,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042700730896331606,
      "loss": 4.7784,
      "mean_token_accuracy": 0.22756955325603484,
      "num_tokens": 59508329.0,
      "step": 32330
    },
    {
      "entropy": 5.07901930809021,
      "epoch": 2.7008853992649513,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004269853588439413,
      "loss": 4.7616,
      "mean_token_accuracy": 0.22497594505548477,
      "num_tokens": 59516799.0,
      "step": 32335
    },
    {
      "entropy": 5.071376514434815,
      "epoch": 2.7013030404276646,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042696340606387977,
      "loss": 4.7058,
      "mean_token_accuracy": 0.2254677027463913,
      "num_tokens": 59526097.0,
      "step": 32340
    },
    {
      "entropy": 5.104973554611206,
      "epoch": 2.701720681590378,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000426941450623516,
      "loss": 4.759,
      "mean_token_accuracy": 0.21984337717294694,
      "num_tokens": 59535287.0,
      "step": 32345
    },
    {
      "entropy": 5.138319301605224,
      "epoch": 2.7021383227530906,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042691949252323427,
      "loss": 4.7004,
      "mean_token_accuracy": 0.22783284187316893,
      "num_tokens": 59545232.0,
      "step": 32350
    },
    {
      "entropy": 4.9977264404296875,
      "epoch": 2.7025559639158034,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004268975317634188,
      "loss": 4.7229,
      "mean_token_accuracy": 0.22783612906932832,
      "num_tokens": 59553977.0,
      "step": 32355
    },
    {
      "entropy": 4.971862316131592,
      "epoch": 2.7029736050785167,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042687556834445435,
      "loss": 4.7431,
      "mean_token_accuracy": 0.22548855990171432,
      "num_tokens": 59563027.0,
      "step": 32360
    },
    {
      "entropy": 4.974747371673584,
      "epoch": 2.7033912462412295,
      "grad_norm": 1.25,
      "learning_rate": 0.0004268536022667252,
      "loss": 4.5897,
      "mean_token_accuracy": 0.2350999727845192,
      "num_tokens": 59570943.0,
      "step": 32365
    },
    {
      "entropy": 5.0670393943786625,
      "epoch": 2.7038088874039428,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042683163353061595,
      "loss": 4.7877,
      "mean_token_accuracy": 0.2236349493265152,
      "num_tokens": 59581323.0,
      "step": 32370
    },
    {
      "entropy": 4.967481517791748,
      "epoch": 2.7042265285666556,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042680966213651126,
      "loss": 4.67,
      "mean_token_accuracy": 0.22829890698194505,
      "num_tokens": 59590429.0,
      "step": 32375
    },
    {
      "entropy": 4.879568576812744,
      "epoch": 2.7046441697293684,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004267876880847956,
      "loss": 4.4632,
      "mean_token_accuracy": 0.2540003746747971,
      "num_tokens": 59598547.0,
      "step": 32380
    },
    {
      "entropy": 5.049269723892212,
      "epoch": 2.7050618108920816,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004267657113758538,
      "loss": 4.6584,
      "mean_token_accuracy": 0.22633178532123566,
      "num_tokens": 59607807.0,
      "step": 32385
    },
    {
      "entropy": 5.004770278930664,
      "epoch": 2.7054794520547945,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042674373201007055,
      "loss": 4.706,
      "mean_token_accuracy": 0.23150249272584916,
      "num_tokens": 59616591.0,
      "step": 32390
    },
    {
      "entropy": 5.056351661682129,
      "epoch": 2.7058970932175077,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004267217499878306,
      "loss": 4.6757,
      "mean_token_accuracy": 0.23465877920389175,
      "num_tokens": 59625162.0,
      "step": 32395
    },
    {
      "entropy": 5.046062088012695,
      "epoch": 2.7063147343802205,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042669976530951883,
      "loss": 4.6459,
      "mean_token_accuracy": 0.23583236783742906,
      "num_tokens": 59633776.0,
      "step": 32400
    },
    {
      "entropy": 5.030474996566772,
      "epoch": 2.7067323755429333,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042667777797552003,
      "loss": 4.6956,
      "mean_token_accuracy": 0.22158671915531158,
      "num_tokens": 59642439.0,
      "step": 32405
    },
    {
      "entropy": 5.014709424972534,
      "epoch": 2.7071500167056466,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042665578798621915,
      "loss": 4.6658,
      "mean_token_accuracy": 0.22871823012828826,
      "num_tokens": 59651241.0,
      "step": 32410
    },
    {
      "entropy": 5.086176109313965,
      "epoch": 2.7075676578683594,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042663379534200113,
      "loss": 4.7804,
      "mean_token_accuracy": 0.21800240874290466,
      "num_tokens": 59660577.0,
      "step": 32415
    },
    {
      "entropy": 4.972119855880737,
      "epoch": 2.7079852990310727,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004266118000432511,
      "loss": 4.6595,
      "mean_token_accuracy": 0.23653375506401061,
      "num_tokens": 59669256.0,
      "step": 32420
    },
    {
      "entropy": 5.017493104934692,
      "epoch": 2.7084029401937855,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00042658980209035394,
      "loss": 4.6685,
      "mean_token_accuracy": 0.234796179831028,
      "num_tokens": 59678240.0,
      "step": 32425
    },
    {
      "entropy": 5.113266086578369,
      "epoch": 2.7088205813564983,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042656780148369477,
      "loss": 4.8649,
      "mean_token_accuracy": 0.2141604870557785,
      "num_tokens": 59688519.0,
      "step": 32430
    },
    {
      "entropy": 5.1213963508605955,
      "epoch": 2.7092382225192115,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004265457982236589,
      "loss": 4.7362,
      "mean_token_accuracy": 0.2241196945309639,
      "num_tokens": 59697482.0,
      "step": 32435
    },
    {
      "entropy": 5.097848510742187,
      "epoch": 2.7096558636819243,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004265237923106314,
      "loss": 4.7188,
      "mean_token_accuracy": 0.22778226733207702,
      "num_tokens": 59706893.0,
      "step": 32440
    },
    {
      "entropy": 4.980399751663208,
      "epoch": 2.7100735048446376,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042650178374499753,
      "loss": 4.5909,
      "mean_token_accuracy": 0.23967147171497344,
      "num_tokens": 59715678.0,
      "step": 32445
    },
    {
      "entropy": 5.0052080154418945,
      "epoch": 2.7104911460073504,
      "grad_norm": 1.015625,
      "learning_rate": 0.00042647977252714247,
      "loss": 4.6126,
      "mean_token_accuracy": 0.2408186599612236,
      "num_tokens": 59724818.0,
      "step": 32450
    },
    {
      "entropy": 4.891044807434082,
      "epoch": 2.7109087871700632,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004264577586574518,
      "loss": 4.5717,
      "mean_token_accuracy": 0.2323321670293808,
      "num_tokens": 59733341.0,
      "step": 32455
    },
    {
      "entropy": 5.0498785972595215,
      "epoch": 2.7113264283327765,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00042643574213631065,
      "loss": 4.7642,
      "mean_token_accuracy": 0.22435592412948607,
      "num_tokens": 59742504.0,
      "step": 32460
    },
    {
      "entropy": 5.088817596435547,
      "epoch": 2.7117440694954897,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004264137229641046,
      "loss": 4.6362,
      "mean_token_accuracy": 0.24085381925106047,
      "num_tokens": 59751749.0,
      "step": 32465
    },
    {
      "entropy": 5.074612045288086,
      "epoch": 2.7121617106582026,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000426391701141219,
      "loss": 4.7384,
      "mean_token_accuracy": 0.22550398856401443,
      "num_tokens": 59761045.0,
      "step": 32470
    },
    {
      "entropy": 4.988625001907349,
      "epoch": 2.7125793518209154,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042636967666803954,
      "loss": 4.7393,
      "mean_token_accuracy": 0.2306083157658577,
      "num_tokens": 59770147.0,
      "step": 32475
    },
    {
      "entropy": 5.0094531059265135,
      "epoch": 2.7129969929836286,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042634764954495167,
      "loss": 4.6445,
      "mean_token_accuracy": 0.2289728879928589,
      "num_tokens": 59779624.0,
      "step": 32480
    },
    {
      "entropy": 5.184315538406372,
      "epoch": 2.7134146341463414,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042632561977234095,
      "loss": 4.9001,
      "mean_token_accuracy": 0.21513957530260086,
      "num_tokens": 59789366.0,
      "step": 32485
    },
    {
      "entropy": 5.049900102615356,
      "epoch": 2.7138322753090547,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004263035873505932,
      "loss": 4.6657,
      "mean_token_accuracy": 0.22627755999565125,
      "num_tokens": 59798043.0,
      "step": 32490
    },
    {
      "entropy": 5.0420349597930905,
      "epoch": 2.7142499164717675,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004262815522800939,
      "loss": 4.7149,
      "mean_token_accuracy": 0.22750913351774216,
      "num_tokens": 59807207.0,
      "step": 32495
    },
    {
      "entropy": 4.981714200973511,
      "epoch": 2.7146675576344803,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004262595145612291,
      "loss": 4.5573,
      "mean_token_accuracy": 0.23933956921100616,
      "num_tokens": 59817004.0,
      "step": 32500
    },
    {
      "entropy": 5.053243160247803,
      "epoch": 2.7150851987971936,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042623747419438427,
      "loss": 4.7057,
      "mean_token_accuracy": 0.22774894684553146,
      "num_tokens": 59825714.0,
      "step": 32505
    },
    {
      "entropy": 5.075837182998657,
      "epoch": 2.7155028399599064,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004262154311799455,
      "loss": 4.7072,
      "mean_token_accuracy": 0.2268832102417946,
      "num_tokens": 59834410.0,
      "step": 32510
    },
    {
      "entropy": 5.056051111221313,
      "epoch": 2.7159204811226196,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004261933855182985,
      "loss": 4.7153,
      "mean_token_accuracy": 0.22583217173814774,
      "num_tokens": 59843629.0,
      "step": 32515
    },
    {
      "entropy": 5.068230009078979,
      "epoch": 2.7163381222853324,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042617133720982943,
      "loss": 4.7957,
      "mean_token_accuracy": 0.2251737892627716,
      "num_tokens": 59852902.0,
      "step": 32520
    },
    {
      "entropy": 5.1157142162323,
      "epoch": 2.7167557634480453,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042614928625492397,
      "loss": 4.7714,
      "mean_token_accuracy": 0.22119743525981903,
      "num_tokens": 59861434.0,
      "step": 32525
    },
    {
      "entropy": 5.012769317626953,
      "epoch": 2.7171734046107585,
      "grad_norm": 1.015625,
      "learning_rate": 0.00042612723265396846,
      "loss": 4.6965,
      "mean_token_accuracy": 0.22547978460788726,
      "num_tokens": 59871591.0,
      "step": 32530
    },
    {
      "entropy": 4.911831712722778,
      "epoch": 2.7175910457734713,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004261051764073488,
      "loss": 4.5768,
      "mean_token_accuracy": 0.23684237152338028,
      "num_tokens": 59880615.0,
      "step": 32535
    },
    {
      "entropy": 5.034332323074341,
      "epoch": 2.7180086869361846,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00042608311751545115,
      "loss": 4.7889,
      "mean_token_accuracy": 0.22683415561914444,
      "num_tokens": 59889558.0,
      "step": 32540
    },
    {
      "entropy": 5.088851404190064,
      "epoch": 2.7184263280988974,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042606105597866163,
      "loss": 4.7078,
      "mean_token_accuracy": 0.2306712418794632,
      "num_tokens": 59897794.0,
      "step": 32545
    },
    {
      "entropy": 5.07924337387085,
      "epoch": 2.71884396926161,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004260389917973665,
      "loss": 4.7793,
      "mean_token_accuracy": 0.22094756513834,
      "num_tokens": 59908074.0,
      "step": 32550
    },
    {
      "entropy": 5.004054307937622,
      "epoch": 2.7192616104243235,
      "grad_norm": 1.09375,
      "learning_rate": 0.000426016924971952,
      "loss": 4.6694,
      "mean_token_accuracy": 0.22763351798057557,
      "num_tokens": 59916986.0,
      "step": 32555
    },
    {
      "entropy": 4.975055313110351,
      "epoch": 2.7196792515870363,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042599485550280446,
      "loss": 4.636,
      "mean_token_accuracy": 0.24021464735269546,
      "num_tokens": 59925539.0,
      "step": 32560
    },
    {
      "entropy": 5.050344562530517,
      "epoch": 2.7200968927497495,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042597278339031024,
      "loss": 4.672,
      "mean_token_accuracy": 0.23016842901706697,
      "num_tokens": 59933732.0,
      "step": 32565
    },
    {
      "entropy": 4.984659671783447,
      "epoch": 2.7205145339124623,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004259507086348557,
      "loss": 4.6395,
      "mean_token_accuracy": 0.23637499362230302,
      "num_tokens": 59942217.0,
      "step": 32570
    },
    {
      "entropy": 5.103958034515381,
      "epoch": 2.720932175075175,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042592863123682735,
      "loss": 4.7825,
      "mean_token_accuracy": 0.22520557790994644,
      "num_tokens": 59951750.0,
      "step": 32575
    },
    {
      "entropy": 5.093322515487671,
      "epoch": 2.7213498162378884,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004259065511966116,
      "loss": 4.793,
      "mean_token_accuracy": 0.219590625166893,
      "num_tokens": 59961054.0,
      "step": 32580
    },
    {
      "entropy": 5.0534059524536135,
      "epoch": 2.7217674574006017,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000425884468514595,
      "loss": 4.759,
      "mean_token_accuracy": 0.21374405920505524,
      "num_tokens": 59971345.0,
      "step": 32585
    },
    {
      "entropy": 5.022346067428589,
      "epoch": 2.7221850985633145,
      "grad_norm": 1.359375,
      "learning_rate": 0.00042586238319116423,
      "loss": 4.6514,
      "mean_token_accuracy": 0.23746173977851867,
      "num_tokens": 59980919.0,
      "step": 32590
    },
    {
      "entropy": 5.089340734481811,
      "epoch": 2.7226027397260273,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042584029522670574,
      "loss": 4.6759,
      "mean_token_accuracy": 0.23396458178758622,
      "num_tokens": 59990881.0,
      "step": 32595
    },
    {
      "entropy": 5.096538972854614,
      "epoch": 2.7230203808887405,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004258182046216064,
      "loss": 4.81,
      "mean_token_accuracy": 0.22506069540977477,
      "num_tokens": 60000556.0,
      "step": 32600
    },
    {
      "entropy": 5.069505071640014,
      "epoch": 2.7234380220514534,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004257961113762528,
      "loss": 4.7309,
      "mean_token_accuracy": 0.22701556235551834,
      "num_tokens": 60009674.0,
      "step": 32605
    },
    {
      "entropy": 5.039439725875854,
      "epoch": 2.7238556632141666,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004257740154910318,
      "loss": 4.735,
      "mean_token_accuracy": 0.21969154328107834,
      "num_tokens": 60019915.0,
      "step": 32610
    },
    {
      "entropy": 5.088975191116333,
      "epoch": 2.7242733043768794,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004257519169663302,
      "loss": 4.7498,
      "mean_token_accuracy": 0.22253550291061402,
      "num_tokens": 60029384.0,
      "step": 32615
    },
    {
      "entropy": 5.059198951721191,
      "epoch": 2.7246909455395922,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042572981580253475,
      "loss": 4.7309,
      "mean_token_accuracy": 0.22239338457584382,
      "num_tokens": 60038763.0,
      "step": 32620
    },
    {
      "entropy": 5.077136993408203,
      "epoch": 2.7251085867023055,
      "grad_norm": 1.046875,
      "learning_rate": 0.00042570771200003255,
      "loss": 4.6852,
      "mean_token_accuracy": 0.2238313615322113,
      "num_tokens": 60048131.0,
      "step": 32625
    },
    {
      "entropy": 4.996273183822632,
      "epoch": 2.7255262278650183,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042568560555921033,
      "loss": 4.6464,
      "mean_token_accuracy": 0.22990435510873794,
      "num_tokens": 60056962.0,
      "step": 32630
    },
    {
      "entropy": 5.029724025726319,
      "epoch": 2.7259438690277316,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004256634964804553,
      "loss": 4.7051,
      "mean_token_accuracy": 0.2311662867665291,
      "num_tokens": 60066818.0,
      "step": 32635
    },
    {
      "entropy": 5.072529745101929,
      "epoch": 2.7263615101904444,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042564138476415433,
      "loss": 4.7073,
      "mean_token_accuracy": 0.22236282825469972,
      "num_tokens": 60075635.0,
      "step": 32640
    },
    {
      "entropy": 5.10526442527771,
      "epoch": 2.726779151353157,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042561927041069467,
      "loss": 4.7748,
      "mean_token_accuracy": 0.22608171701431273,
      "num_tokens": 60084531.0,
      "step": 32645
    },
    {
      "entropy": 5.061924457550049,
      "epoch": 2.7271967925158704,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042559715342046334,
      "loss": 4.7174,
      "mean_token_accuracy": 0.22387922704219818,
      "num_tokens": 60092992.0,
      "step": 32650
    },
    {
      "entropy": 5.016059684753418,
      "epoch": 2.7276144336785833,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004255750337938475,
      "loss": 4.6573,
      "mean_token_accuracy": 0.23266577273607253,
      "num_tokens": 60101804.0,
      "step": 32655
    },
    {
      "entropy": 5.088020849227905,
      "epoch": 2.7280320748412965,
      "grad_norm": 1.015625,
      "learning_rate": 0.00042555291153123455,
      "loss": 4.7546,
      "mean_token_accuracy": 0.21893056780099868,
      "num_tokens": 60111417.0,
      "step": 32660
    },
    {
      "entropy": 5.031080722808838,
      "epoch": 2.7284497160040093,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042553078663301163,
      "loss": 4.6269,
      "mean_token_accuracy": 0.23317282795906066,
      "num_tokens": 60120154.0,
      "step": 32665
    },
    {
      "entropy": 5.124962234497071,
      "epoch": 2.728867357166722,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042550865909956606,
      "loss": 4.783,
      "mean_token_accuracy": 0.2179592415690422,
      "num_tokens": 60128762.0,
      "step": 32670
    },
    {
      "entropy": 4.966511297225952,
      "epoch": 2.7292849983294354,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042548652893128526,
      "loss": 4.6281,
      "mean_token_accuracy": 0.22981294691562654,
      "num_tokens": 60138091.0,
      "step": 32675
    },
    {
      "entropy": 5.000231599807739,
      "epoch": 2.729702639492148,
      "grad_norm": 1.125,
      "learning_rate": 0.0004254643961285567,
      "loss": 4.6192,
      "mean_token_accuracy": 0.23378575593233109,
      "num_tokens": 60146937.0,
      "step": 32680
    },
    {
      "entropy": 4.993679141998291,
      "epoch": 2.7301202806548615,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00042544226069176765,
      "loss": 4.6565,
      "mean_token_accuracy": 0.23294521272182464,
      "num_tokens": 60155790.0,
      "step": 32685
    },
    {
      "entropy": 4.962065744400024,
      "epoch": 2.7305379218175743,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004254201226213059,
      "loss": 4.6209,
      "mean_token_accuracy": 0.23339169025421141,
      "num_tokens": 60165620.0,
      "step": 32690
    },
    {
      "entropy": 5.028203296661377,
      "epoch": 2.730955562980287,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004253979819175587,
      "loss": 4.6433,
      "mean_token_accuracy": 0.23360038697719573,
      "num_tokens": 60175544.0,
      "step": 32695
    },
    {
      "entropy": 5.103120803833008,
      "epoch": 2.7313732041430003,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042537583858091383,
      "loss": 4.7977,
      "mean_token_accuracy": 0.21939945071935654,
      "num_tokens": 60184177.0,
      "step": 32700
    },
    {
      "entropy": 4.962979316711426,
      "epoch": 2.7317908453057136,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004253536926117589,
      "loss": 4.61,
      "mean_token_accuracy": 0.23609544038772584,
      "num_tokens": 60192417.0,
      "step": 32705
    },
    {
      "entropy": 5.093804979324341,
      "epoch": 2.7322084864684264,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004253315440104816,
      "loss": 4.771,
      "mean_token_accuracy": 0.22361129373311997,
      "num_tokens": 60201679.0,
      "step": 32710
    },
    {
      "entropy": 5.070327711105347,
      "epoch": 2.732626127631139,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042530939277746964,
      "loss": 4.7101,
      "mean_token_accuracy": 0.23261211961507797,
      "num_tokens": 60210228.0,
      "step": 32715
    },
    {
      "entropy": 4.952836561203003,
      "epoch": 2.7330437687938525,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004252872389131109,
      "loss": 4.582,
      "mean_token_accuracy": 0.2327900528907776,
      "num_tokens": 60218958.0,
      "step": 32720
    },
    {
      "entropy": 4.977156257629394,
      "epoch": 2.7334614099565653,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004252650824177931,
      "loss": 4.6549,
      "mean_token_accuracy": 0.22810798585414888,
      "num_tokens": 60228490.0,
      "step": 32725
    },
    {
      "entropy": 5.0794189929962155,
      "epoch": 2.7338790511192785,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004252429232919042,
      "loss": 4.7703,
      "mean_token_accuracy": 0.22589764893054962,
      "num_tokens": 60238289.0,
      "step": 32730
    },
    {
      "entropy": 4.974723243713379,
      "epoch": 2.7342966922819913,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000425220761535832,
      "loss": 4.5763,
      "mean_token_accuracy": 0.23025186210870743,
      "num_tokens": 60247731.0,
      "step": 32735
    },
    {
      "entropy": 5.150882053375244,
      "epoch": 2.734714333444704,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004251985971499647,
      "loss": 4.8723,
      "mean_token_accuracy": 0.22029241770505906,
      "num_tokens": 60257073.0,
      "step": 32740
    },
    {
      "entropy": 4.986446857452393,
      "epoch": 2.7351319746074174,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042517643013469,
      "loss": 4.5616,
      "mean_token_accuracy": 0.24242764115333557,
      "num_tokens": 60265354.0,
      "step": 32745
    },
    {
      "entropy": 5.069497537612915,
      "epoch": 2.7355496157701302,
      "grad_norm": 1.0,
      "learning_rate": 0.0004251542604903962,
      "loss": 4.7599,
      "mean_token_accuracy": 0.21482582986354828,
      "num_tokens": 60275456.0,
      "step": 32750
    },
    {
      "entropy": 5.0873942375183105,
      "epoch": 2.7359672569328435,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004251320882174713,
      "loss": 4.8233,
      "mean_token_accuracy": 0.2148478865623474,
      "num_tokens": 60284051.0,
      "step": 32755
    },
    {
      "entropy": 5.052629089355468,
      "epoch": 2.7363848980955563,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004251099133163035,
      "loss": 4.6501,
      "mean_token_accuracy": 0.2251687914133072,
      "num_tokens": 60293590.0,
      "step": 32760
    },
    {
      "entropy": 5.065043163299561,
      "epoch": 2.736802539258269,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000425087735787281,
      "loss": 4.6589,
      "mean_token_accuracy": 0.2312410831451416,
      "num_tokens": 60302107.0,
      "step": 32765
    },
    {
      "entropy": 5.037451171875,
      "epoch": 2.7372201804209824,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042506555563079194,
      "loss": 4.7246,
      "mean_token_accuracy": 0.23135038167238237,
      "num_tokens": 60311212.0,
      "step": 32770
    },
    {
      "entropy": 5.066587829589844,
      "epoch": 2.737637821583695,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004250433728472248,
      "loss": 4.7044,
      "mean_token_accuracy": 0.2266307070851326,
      "num_tokens": 60320282.0,
      "step": 32775
    },
    {
      "entropy": 5.11245551109314,
      "epoch": 2.7380554627464084,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004250211874369678,
      "loss": 4.6928,
      "mean_token_accuracy": 0.231039197742939,
      "num_tokens": 60328386.0,
      "step": 32780
    },
    {
      "entropy": 5.065380477905274,
      "epoch": 2.7384731039091212,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004249989994004093,
      "loss": 4.7237,
      "mean_token_accuracy": 0.2290617436170578,
      "num_tokens": 60338210.0,
      "step": 32785
    },
    {
      "entropy": 5.017452383041382,
      "epoch": 2.738890745071834,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042497680873793776,
      "loss": 4.675,
      "mean_token_accuracy": 0.2357072800397873,
      "num_tokens": 60347347.0,
      "step": 32790
    },
    {
      "entropy": 5.091592264175415,
      "epoch": 2.7393083862345473,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004249546154499416,
      "loss": 4.7634,
      "mean_token_accuracy": 0.22303192764520646,
      "num_tokens": 60356211.0,
      "step": 32795
    },
    {
      "entropy": 5.08235182762146,
      "epoch": 2.73972602739726,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042493241953680947,
      "loss": 4.7582,
      "mean_token_accuracy": 0.22431036233901977,
      "num_tokens": 60365678.0,
      "step": 32800
    },
    {
      "entropy": 5.071024751663208,
      "epoch": 2.7401436685599734,
      "grad_norm": 1.125,
      "learning_rate": 0.00042491022099892985,
      "loss": 4.7161,
      "mean_token_accuracy": 0.22175025939941406,
      "num_tokens": 60375243.0,
      "step": 32805
    },
    {
      "entropy": 5.019820928573608,
      "epoch": 2.740561309722686,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004248880198366913,
      "loss": 4.7398,
      "mean_token_accuracy": 0.22532511204481126,
      "num_tokens": 60385087.0,
      "step": 32810
    },
    {
      "entropy": 5.040626859664917,
      "epoch": 2.740978950885399,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004248658160504827,
      "loss": 4.7661,
      "mean_token_accuracy": 0.2303236871957779,
      "num_tokens": 60394340.0,
      "step": 32815
    },
    {
      "entropy": 5.117239856719971,
      "epoch": 2.7413965920481123,
      "grad_norm": 1.125,
      "learning_rate": 0.00042484360964069247,
      "loss": 4.7236,
      "mean_token_accuracy": 0.22454709261655809,
      "num_tokens": 60404537.0,
      "step": 32820
    },
    {
      "entropy": 4.926508951187134,
      "epoch": 2.7418142332108255,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004248214006077095,
      "loss": 4.5735,
      "mean_token_accuracy": 0.23548229336738585,
      "num_tokens": 60414124.0,
      "step": 32825
    },
    {
      "entropy": 4.913050699234009,
      "epoch": 2.7422318743735383,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00042479918895192253,
      "loss": 4.553,
      "mean_token_accuracy": 0.23790228962898255,
      "num_tokens": 60422719.0,
      "step": 32830
    },
    {
      "entropy": 4.968108034133911,
      "epoch": 2.742649515536251,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042477697467372056,
      "loss": 4.6033,
      "mean_token_accuracy": 0.23906211853027343,
      "num_tokens": 60430877.0,
      "step": 32835
    },
    {
      "entropy": 5.004045104980468,
      "epoch": 2.7430671566989644,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00042475475777349234,
      "loss": 4.696,
      "mean_token_accuracy": 0.23068926334381104,
      "num_tokens": 60440320.0,
      "step": 32840
    },
    {
      "entropy": 4.9645383834838865,
      "epoch": 2.743484797861677,
      "grad_norm": 1.046875,
      "learning_rate": 0.00042473253825162674,
      "loss": 4.6049,
      "mean_token_accuracy": 0.23282813727855683,
      "num_tokens": 60449442.0,
      "step": 32845
    },
    {
      "entropy": 4.996943473815918,
      "epoch": 2.7439024390243905,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004247103161085129,
      "loss": 4.6185,
      "mean_token_accuracy": 0.23742079436779023,
      "num_tokens": 60458176.0,
      "step": 32850
    },
    {
      "entropy": 5.00455322265625,
      "epoch": 2.7443200801871033,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042468809134453973,
      "loss": 4.6741,
      "mean_token_accuracy": 0.2307494103908539,
      "num_tokens": 60466842.0,
      "step": 32855
    },
    {
      "entropy": 4.985470724105835,
      "epoch": 2.744737721349816,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004246658639600963,
      "loss": 4.6645,
      "mean_token_accuracy": 0.23206437528133392,
      "num_tokens": 60475591.0,
      "step": 32860
    },
    {
      "entropy": 4.980706644058228,
      "epoch": 2.7451553625125293,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004246436339555719,
      "loss": 4.6362,
      "mean_token_accuracy": 0.2323045775294304,
      "num_tokens": 60485263.0,
      "step": 32865
    },
    {
      "entropy": 5.0205769538879395,
      "epoch": 2.745573003675242,
      "grad_norm": 1.046875,
      "learning_rate": 0.00042462140133135544,
      "loss": 4.6732,
      "mean_token_accuracy": 0.23207363933324815,
      "num_tokens": 60494256.0,
      "step": 32870
    },
    {
      "entropy": 4.897966527938843,
      "epoch": 2.7459906448379554,
      "grad_norm": 1.125,
      "learning_rate": 0.0004245991660878363,
      "loss": 4.5503,
      "mean_token_accuracy": 0.23935185968875886,
      "num_tokens": 60502920.0,
      "step": 32875
    },
    {
      "entropy": 4.972193336486816,
      "epoch": 2.746408286000668,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004245769282254037,
      "loss": 4.5901,
      "mean_token_accuracy": 0.23896284103393556,
      "num_tokens": 60511827.0,
      "step": 32880
    },
    {
      "entropy": 5.10370569229126,
      "epoch": 2.746825927163381,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042455468774444695,
      "loss": 4.7649,
      "mean_token_accuracy": 0.2271669551730156,
      "num_tokens": 60521602.0,
      "step": 32885
    },
    {
      "entropy": 4.918258476257324,
      "epoch": 2.7472435683260943,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042453244464535524,
      "loss": 4.5423,
      "mean_token_accuracy": 0.23798671662807463,
      "num_tokens": 60531016.0,
      "step": 32890
    },
    {
      "entropy": 4.958998537063598,
      "epoch": 2.747661209488807,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004245101989285182,
      "loss": 4.6585,
      "mean_token_accuracy": 0.22826471775770188,
      "num_tokens": 60540112.0,
      "step": 32895
    },
    {
      "entropy": 4.985217094421387,
      "epoch": 2.7480788506515204,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004244879505943252,
      "loss": 4.6719,
      "mean_token_accuracy": 0.23169296979904175,
      "num_tokens": 60549577.0,
      "step": 32900
    },
    {
      "entropy": 5.099730014801025,
      "epoch": 2.748496491814233,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00042446569964316557,
      "loss": 4.7455,
      "mean_token_accuracy": 0.22412047535181046,
      "num_tokens": 60558874.0,
      "step": 32905
    },
    {
      "entropy": 5.054090309143066,
      "epoch": 2.748914132976946,
      "grad_norm": 1.140625,
      "learning_rate": 0.000424443446075429,
      "loss": 4.7026,
      "mean_token_accuracy": 0.22729025781154633,
      "num_tokens": 60568252.0,
      "step": 32910
    },
    {
      "entropy": 4.9750871658325195,
      "epoch": 2.7493317741396592,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004244211898915051,
      "loss": 4.5469,
      "mean_token_accuracy": 0.2349572405219078,
      "num_tokens": 60577184.0,
      "step": 32915
    },
    {
      "entropy": 4.973816013336181,
      "epoch": 2.749749415302372,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004243989310917833,
      "loss": 4.6406,
      "mean_token_accuracy": 0.23091114461421966,
      "num_tokens": 60585679.0,
      "step": 32920
    },
    {
      "entropy": 4.999643802642822,
      "epoch": 2.7501670564650853,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042437666967665345,
      "loss": 4.7173,
      "mean_token_accuracy": 0.2280352532863617,
      "num_tokens": 60595010.0,
      "step": 32925
    },
    {
      "entropy": 5.057349061965942,
      "epoch": 2.750584697627798,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00042435440564650526,
      "loss": 4.6702,
      "mean_token_accuracy": 0.23126665949821473,
      "num_tokens": 60604240.0,
      "step": 32930
    },
    {
      "entropy": 5.038579511642456,
      "epoch": 2.751002338790511,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004243321390017283,
      "loss": 4.7671,
      "mean_token_accuracy": 0.22289977222681046,
      "num_tokens": 60613817.0,
      "step": 32935
    },
    {
      "entropy": 5.007104396820068,
      "epoch": 2.751419979953224,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004243098697427125,
      "loss": 4.6361,
      "mean_token_accuracy": 0.23135066777467728,
      "num_tokens": 60622366.0,
      "step": 32940
    },
    {
      "entropy": 4.988522291183472,
      "epoch": 2.7518376211159374,
      "grad_norm": 1.03125,
      "learning_rate": 0.00042428759786984773,
      "loss": 4.6998,
      "mean_token_accuracy": 0.23283661901950836,
      "num_tokens": 60630906.0,
      "step": 32945
    },
    {
      "entropy": 4.976337671279907,
      "epoch": 2.7522552622786502,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004242653233835239,
      "loss": 4.6149,
      "mean_token_accuracy": 0.23565127700567245,
      "num_tokens": 60640920.0,
      "step": 32950
    },
    {
      "entropy": 4.991847658157349,
      "epoch": 2.752672903441363,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042424304628413086,
      "loss": 4.667,
      "mean_token_accuracy": 0.22915260940790178,
      "num_tokens": 60650194.0,
      "step": 32955
    },
    {
      "entropy": 5.111684131622314,
      "epoch": 2.7530905446040763,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004242207665720587,
      "loss": 4.8307,
      "mean_token_accuracy": 0.21546273529529572,
      "num_tokens": 60659824.0,
      "step": 32960
    },
    {
      "entropy": 5.0393486499786375,
      "epoch": 2.753508185766789,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042419848424769737,
      "loss": 4.6814,
      "mean_token_accuracy": 0.22869132757186889,
      "num_tokens": 60669298.0,
      "step": 32965
    },
    {
      "entropy": 5.016578483581543,
      "epoch": 2.7539258269295024,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000424176199311437,
      "loss": 4.6918,
      "mean_token_accuracy": 0.22794285118579866,
      "num_tokens": 60679491.0,
      "step": 32970
    },
    {
      "entropy": 5.0825310230255125,
      "epoch": 2.754343468092215,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042415391176366763,
      "loss": 4.6927,
      "mean_token_accuracy": 0.21906275302171707,
      "num_tokens": 60688975.0,
      "step": 32975
    },
    {
      "entropy": 5.0064610004425045,
      "epoch": 2.754761109254928,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004241316216047796,
      "loss": 4.6175,
      "mean_token_accuracy": 0.2261264830827713,
      "num_tokens": 60697914.0,
      "step": 32980
    },
    {
      "entropy": 4.962593460083008,
      "epoch": 2.7551787504176413,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042410932883516297,
      "loss": 4.5979,
      "mean_token_accuracy": 0.2394718423485756,
      "num_tokens": 60706690.0,
      "step": 32985
    },
    {
      "entropy": 4.954294681549072,
      "epoch": 2.755596391580354,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004240870334552081,
      "loss": 4.7368,
      "mean_token_accuracy": 0.22595674991607667,
      "num_tokens": 60715377.0,
      "step": 32990
    },
    {
      "entropy": 5.06918044090271,
      "epoch": 2.7560140327430673,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042406473546530523,
      "loss": 4.7033,
      "mean_token_accuracy": 0.2316712960600853,
      "num_tokens": 60724021.0,
      "step": 32995
    },
    {
      "entropy": 5.083750295639038,
      "epoch": 2.75643167390578,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00042404243486584463,
      "loss": 4.7676,
      "mean_token_accuracy": 0.22659618258476258,
      "num_tokens": 60733174.0,
      "step": 33000
    },
    {
      "epoch": 2.75643167390578,
      "eval_entropy": 4.805341296568686,
      "eval_loss": 4.724085807800293,
      "eval_mean_token_accuracy": 0.23785885811495946,
      "eval_num_tokens": 60733174.0,
      "eval_runtime": 27.5299,
      "eval_samples_per_second": 1371.709,
      "eval_steps_per_second": 171.486,
      "step": 33000
    }
  ],
  "logging_steps": 5,
  "max_steps": 119720,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 3000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 8.9872219109376e+16,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}