{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.7842076798269337,
  "eval_steps": 500,
  "global_step": 3500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 10.742581367492676,
      "epoch": 0.005408328826392645,
      "grad_norm": 5.4375,
      "learning_rate": 2e-06,
      "loss": 10.7836,
      "mean_token_accuracy": 0.0,
      "num_tokens": 10942.0,
      "step": 5
    },
    {
      "entropy": 10.742576122283936,
      "epoch": 0.01081665765278529,
      "grad_norm": 5.0,
      "learning_rate": 4.5e-06,
      "loss": 10.7725,
      "mean_token_accuracy": 0.0,
      "num_tokens": 21887.0,
      "step": 10
    },
    {
      "entropy": 10.742580318450928,
      "epoch": 0.016224986479177934,
      "grad_norm": 5.46875,
      "learning_rate": 7e-06,
      "loss": 10.7688,
      "mean_token_accuracy": 0.0,
      "num_tokens": 34345.0,
      "step": 15
    },
    {
      "entropy": 10.74259843826294,
      "epoch": 0.02163331530557058,
      "grad_norm": 5.3125,
      "learning_rate": 9.5e-06,
      "loss": 10.6831,
      "mean_token_accuracy": 0.0005558153847232461,
      "num_tokens": 45445.0,
      "step": 20
    },
    {
      "entropy": 10.742469692230225,
      "epoch": 0.02704164413196322,
      "grad_norm": 4.34375,
      "learning_rate": 1.2e-05,
      "loss": 10.5988,
      "mean_token_accuracy": 0.007499191560782492,
      "num_tokens": 58237.0,
      "step": 25
    },
    {
      "entropy": 10.742159843444824,
      "epoch": 0.03244997295835587,
      "grad_norm": 4.25,
      "learning_rate": 1.4500000000000002e-05,
      "loss": 10.4088,
      "mean_token_accuracy": 0.031410138495266436,
      "num_tokens": 68098.0,
      "step": 30
    },
    {
      "entropy": 10.740957832336425,
      "epoch": 0.03785830178474851,
      "grad_norm": 3.203125,
      "learning_rate": 1.7000000000000003e-05,
      "loss": 10.2937,
      "mean_token_accuracy": 0.04090985879302025,
      "num_tokens": 79867.0,
      "step": 35
    },
    {
      "entropy": 10.738431835174561,
      "epoch": 0.04326663061114116,
      "grad_norm": 2.625,
      "learning_rate": 1.95e-05,
      "loss": 10.1592,
      "mean_token_accuracy": 0.037697183154523375,
      "num_tokens": 91257.0,
      "step": 40
    },
    {
      "entropy": 10.735397911071777,
      "epoch": 0.048674959437533805,
      "grad_norm": 2.1875,
      "learning_rate": 2.2e-05,
      "loss": 10.0705,
      "mean_token_accuracy": 0.04106438439339399,
      "num_tokens": 103474.0,
      "step": 45
    },
    {
      "entropy": 10.733191680908202,
      "epoch": 0.05408328826392644,
      "grad_norm": 2.28125,
      "learning_rate": 2.4500000000000003e-05,
      "loss": 9.9433,
      "mean_token_accuracy": 0.037430111691355704,
      "num_tokens": 114830.0,
      "step": 50
    },
    {
      "entropy": 10.732486057281495,
      "epoch": 0.05949161709031909,
      "grad_norm": 2.046875,
      "learning_rate": 2.7e-05,
      "loss": 9.8905,
      "mean_token_accuracy": 0.036898509971797465,
      "num_tokens": 126017.0,
      "step": 55
    },
    {
      "entropy": 10.73141107559204,
      "epoch": 0.06489994591671173,
      "grad_norm": 2.03125,
      "learning_rate": 2.95e-05,
      "loss": 9.8185,
      "mean_token_accuracy": 0.04504128023982048,
      "num_tokens": 138509.0,
      "step": 60
    },
    {
      "entropy": 10.729285430908202,
      "epoch": 0.07030827474310439,
      "grad_norm": 1.96875,
      "learning_rate": 3.2e-05,
      "loss": 9.764,
      "mean_token_accuracy": 0.0458514004945755,
      "num_tokens": 148353.0,
      "step": 65
    },
    {
      "entropy": 10.726380920410156,
      "epoch": 0.07571660356949703,
      "grad_norm": 1.9140625,
      "learning_rate": 3.4500000000000005e-05,
      "loss": 9.7481,
      "mean_token_accuracy": 0.04072931818664074,
      "num_tokens": 160269.0,
      "step": 70
    },
    {
      "entropy": 10.723598289489747,
      "epoch": 0.08112493239588967,
      "grad_norm": 2.0,
      "learning_rate": 3.7e-05,
      "loss": 9.6457,
      "mean_token_accuracy": 0.040256797894835473,
      "num_tokens": 171568.0,
      "step": 75
    },
    {
      "entropy": 10.720184516906738,
      "epoch": 0.08653326122228232,
      "grad_norm": 2.046875,
      "learning_rate": 3.95e-05,
      "loss": 9.5709,
      "mean_token_accuracy": 0.04319700971245766,
      "num_tokens": 184243.0,
      "step": 80
    },
    {
      "entropy": 10.715407657623292,
      "epoch": 0.09194159004867496,
      "grad_norm": 2.0,
      "learning_rate": 4.2000000000000004e-05,
      "loss": 9.5081,
      "mean_token_accuracy": 0.044286393001675604,
      "num_tokens": 194887.0,
      "step": 85
    },
    {
      "entropy": 10.70880765914917,
      "epoch": 0.09734991887506761,
      "grad_norm": 1.9453125,
      "learning_rate": 4.45e-05,
      "loss": 9.4386,
      "mean_token_accuracy": 0.04779103547334671,
      "num_tokens": 206476.0,
      "step": 90
    },
    {
      "entropy": 10.699271774291992,
      "epoch": 0.10275824770146025,
      "grad_norm": 1.9453125,
      "learning_rate": 4.7000000000000004e-05,
      "loss": 9.3288,
      "mean_token_accuracy": 0.052064205333590505,
      "num_tokens": 216578.0,
      "step": 95
    },
    {
      "entropy": 10.6855544090271,
      "epoch": 0.10816657652785289,
      "grad_norm": 1.90625,
      "learning_rate": 4.9500000000000004e-05,
      "loss": 9.2367,
      "mean_token_accuracy": 0.05270014926791191,
      "num_tokens": 226941.0,
      "step": 100
    },
    {
      "entropy": 10.670534229278564,
      "epoch": 0.11357490535424554,
      "grad_norm": 1.8828125,
      "learning_rate": 5.2e-05,
      "loss": 9.1416,
      "mean_token_accuracy": 0.05506536141037941,
      "num_tokens": 238060.0,
      "step": 105
    },
    {
      "entropy": 10.650165462493897,
      "epoch": 0.11898323418063818,
      "grad_norm": 2.015625,
      "learning_rate": 5.45e-05,
      "loss": 8.998,
      "mean_token_accuracy": 0.06617085449397564,
      "num_tokens": 249137.0,
      "step": 110
    },
    {
      "entropy": 10.614893913269043,
      "epoch": 0.12439156300703083,
      "grad_norm": 1.8984375,
      "learning_rate": 5.7e-05,
      "loss": 8.9011,
      "mean_token_accuracy": 0.0645027995109558,
      "num_tokens": 260712.0,
      "step": 115
    },
    {
      "entropy": 10.56815938949585,
      "epoch": 0.12979989183342347,
      "grad_norm": 1.921875,
      "learning_rate": 5.9499999999999996e-05,
      "loss": 8.7629,
      "mean_token_accuracy": 0.06179735250771046,
      "num_tokens": 271039.0,
      "step": 120
    },
    {
      "entropy": 10.511663913726807,
      "epoch": 0.1352082206598161,
      "grad_norm": 1.984375,
      "learning_rate": 6.2e-05,
      "loss": 8.6203,
      "mean_token_accuracy": 0.0636001892387867,
      "num_tokens": 281353.0,
      "step": 125
    },
    {
      "entropy": 10.432361125946045,
      "epoch": 0.14061654948620878,
      "grad_norm": 1.671875,
      "learning_rate": 6.450000000000001e-05,
      "loss": 8.5429,
      "mean_token_accuracy": 0.07175168134272099,
      "num_tokens": 292221.0,
      "step": 130
    },
    {
      "entropy": 10.352851963043213,
      "epoch": 0.1460248783126014,
      "grad_norm": 1.671875,
      "learning_rate": 6.7e-05,
      "loss": 8.4081,
      "mean_token_accuracy": 0.06957007274031639,
      "num_tokens": 303353.0,
      "step": 135
    },
    {
      "entropy": 10.268811798095703,
      "epoch": 0.15143320713899405,
      "grad_norm": 1.5703125,
      "learning_rate": 6.950000000000001e-05,
      "loss": 8.292,
      "mean_token_accuracy": 0.07166561894118786,
      "num_tokens": 314165.0,
      "step": 140
    },
    {
      "entropy": 10.154801845550537,
      "epoch": 0.1568415359653867,
      "grad_norm": 1.5234375,
      "learning_rate": 7.2e-05,
      "loss": 8.1152,
      "mean_token_accuracy": 0.06947772391140461,
      "num_tokens": 326387.0,
      "step": 145
    },
    {
      "entropy": 10.024849987030029,
      "epoch": 0.16224986479177933,
      "grad_norm": 1.46875,
      "learning_rate": 7.45e-05,
      "loss": 8.0602,
      "mean_token_accuracy": 0.06997468769550323,
      "num_tokens": 339155.0,
      "step": 150
    },
    {
      "entropy": 9.84305181503296,
      "epoch": 0.167658193618172,
      "grad_norm": 1.609375,
      "learning_rate": 7.7e-05,
      "loss": 7.9106,
      "mean_token_accuracy": 0.07246845588088036,
      "num_tokens": 350643.0,
      "step": 155
    },
    {
      "entropy": 9.628917694091797,
      "epoch": 0.17306652244456464,
      "grad_norm": 1.296875,
      "learning_rate": 7.950000000000001e-05,
      "loss": 7.7825,
      "mean_token_accuracy": 0.07223140001296997,
      "num_tokens": 361977.0,
      "step": 160
    },
    {
      "entropy": 9.454836750030518,
      "epoch": 0.17847485127095727,
      "grad_norm": 1.359375,
      "learning_rate": 8.2e-05,
      "loss": 7.7462,
      "mean_token_accuracy": 0.07521784901618958,
      "num_tokens": 373604.0,
      "step": 165
    },
    {
      "entropy": 9.20318956375122,
      "epoch": 0.1838831800973499,
      "grad_norm": 0.98046875,
      "learning_rate": 8.450000000000001e-05,
      "loss": 7.6587,
      "mean_token_accuracy": 0.07736964598298073,
      "num_tokens": 384306.0,
      "step": 170
    },
    {
      "entropy": 8.976179218292236,
      "epoch": 0.18929150892374255,
      "grad_norm": 1.015625,
      "learning_rate": 8.7e-05,
      "loss": 7.5228,
      "mean_token_accuracy": 0.08294685482978821,
      "num_tokens": 396479.0,
      "step": 175
    },
    {
      "entropy": 8.758621406555175,
      "epoch": 0.19469983775013522,
      "grad_norm": 0.95703125,
      "learning_rate": 8.95e-05,
      "loss": 7.4294,
      "mean_token_accuracy": 0.08506134003400803,
      "num_tokens": 408489.0,
      "step": 180
    },
    {
      "entropy": 8.530921840667725,
      "epoch": 0.20010816657652786,
      "grad_norm": 1.0078125,
      "learning_rate": 9.2e-05,
      "loss": 7.3995,
      "mean_token_accuracy": 0.08729378059506417,
      "num_tokens": 419339.0,
      "step": 185
    },
    {
      "entropy": 8.382069301605224,
      "epoch": 0.2055164954029205,
      "grad_norm": 0.9296875,
      "learning_rate": 9.45e-05,
      "loss": 7.3478,
      "mean_token_accuracy": 0.08495231121778488,
      "num_tokens": 430486.0,
      "step": 190
    },
    {
      "entropy": 8.212819290161132,
      "epoch": 0.21092482422931313,
      "grad_norm": 0.92578125,
      "learning_rate": 9.7e-05,
      "loss": 7.3377,
      "mean_token_accuracy": 0.08699661865830421,
      "num_tokens": 440414.0,
      "step": 195
    },
    {
      "entropy": 8.141395664215088,
      "epoch": 0.21633315305570577,
      "grad_norm": 0.84375,
      "learning_rate": 9.95e-05,
      "loss": 7.2746,
      "mean_token_accuracy": 0.0853013277053833,
      "num_tokens": 451475.0,
      "step": 200
    },
    {
      "entropy": 8.020270109176636,
      "epoch": 0.22174148188209844,
      "grad_norm": 0.86328125,
      "learning_rate": 0.000102,
      "loss": 7.3305,
      "mean_token_accuracy": 0.08908139243721962,
      "num_tokens": 462053.0,
      "step": 205
    },
    {
      "entropy": 8.034120750427245,
      "epoch": 0.22714981070849108,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00010449999999999999,
      "loss": 7.2768,
      "mean_token_accuracy": 0.08668714761734009,
      "num_tokens": 473207.0,
      "step": 210
    },
    {
      "entropy": 7.990876579284668,
      "epoch": 0.23255813953488372,
      "grad_norm": 0.9140625,
      "learning_rate": 0.000107,
      "loss": 7.2574,
      "mean_token_accuracy": 0.08839641213417053,
      "num_tokens": 484999.0,
      "step": 215
    },
    {
      "entropy": 7.926925563812256,
      "epoch": 0.23796646836127636,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0001095,
      "loss": 7.2424,
      "mean_token_accuracy": 0.08917610868811607,
      "num_tokens": 496772.0,
      "step": 220
    },
    {
      "entropy": 7.89932656288147,
      "epoch": 0.24337479718766902,
      "grad_norm": 1.0,
      "learning_rate": 0.000112,
      "loss": 7.222,
      "mean_token_accuracy": 0.08830738216638565,
      "num_tokens": 507782.0,
      "step": 225
    },
    {
      "entropy": 7.855933761596679,
      "epoch": 0.24878312601406166,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0001145,
      "loss": 7.1208,
      "mean_token_accuracy": 0.09142142534255981,
      "num_tokens": 518950.0,
      "step": 230
    },
    {
      "entropy": 7.874954509735107,
      "epoch": 0.2541914548404543,
      "grad_norm": 0.984375,
      "learning_rate": 0.00011700000000000001,
      "loss": 7.1844,
      "mean_token_accuracy": 0.09523893147706985,
      "num_tokens": 530795.0,
      "step": 235
    },
    {
      "entropy": 7.853066539764404,
      "epoch": 0.25959978366684694,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00011949999999999999,
      "loss": 7.202,
      "mean_token_accuracy": 0.09066410064697265,
      "num_tokens": 542878.0,
      "step": 240
    },
    {
      "entropy": 7.815495109558105,
      "epoch": 0.2650081124932396,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000122,
      "loss": 7.0976,
      "mean_token_accuracy": 0.09599302783608436,
      "num_tokens": 553682.0,
      "step": 245
    },
    {
      "entropy": 7.718054342269897,
      "epoch": 0.2704164413196322,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0001245,
      "loss": 7.0854,
      "mean_token_accuracy": 0.10266833081841469,
      "num_tokens": 564433.0,
      "step": 250
    },
    {
      "entropy": 7.75045108795166,
      "epoch": 0.27582477014602486,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000127,
      "loss": 7.1247,
      "mean_token_accuracy": 0.09944702386856079,
      "num_tokens": 576001.0,
      "step": 255
    },
    {
      "entropy": 7.712650632858276,
      "epoch": 0.28123309897241755,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0001295,
      "loss": 7.0446,
      "mean_token_accuracy": 0.10567303821444511,
      "num_tokens": 587138.0,
      "step": 260
    },
    {
      "entropy": 7.598682832717896,
      "epoch": 0.2866414277988102,
      "grad_norm": 1.015625,
      "learning_rate": 0.000132,
      "loss": 7.0066,
      "mean_token_accuracy": 0.10336644649505615,
      "num_tokens": 597572.0,
      "step": 265
    },
    {
      "entropy": 7.61051869392395,
      "epoch": 0.2920497566252028,
      "grad_norm": 1.0625,
      "learning_rate": 0.00013450000000000002,
      "loss": 6.9642,
      "mean_token_accuracy": 0.10717165470123291,
      "num_tokens": 608879.0,
      "step": 270
    },
    {
      "entropy": 7.656619739532471,
      "epoch": 0.29745808545159547,
      "grad_norm": 1.21875,
      "learning_rate": 0.00013700000000000002,
      "loss": 7.0022,
      "mean_token_accuracy": 0.09820161312818527,
      "num_tokens": 620439.0,
      "step": 275
    },
    {
      "entropy": 7.668045330047607,
      "epoch": 0.3028664142779881,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0001395,
      "loss": 7.0632,
      "mean_token_accuracy": 0.10111433267593384,
      "num_tokens": 632543.0,
      "step": 280
    },
    {
      "entropy": 7.635419225692749,
      "epoch": 0.30827474310438074,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00014199999999999998,
      "loss": 7.1026,
      "mean_token_accuracy": 0.10522567853331566,
      "num_tokens": 643369.0,
      "step": 285
    },
    {
      "entropy": 7.584413766860962,
      "epoch": 0.3136830719307734,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0001445,
      "loss": 6.9646,
      "mean_token_accuracy": 0.1047127217054367,
      "num_tokens": 655328.0,
      "step": 290
    },
    {
      "entropy": 7.67089467048645,
      "epoch": 0.319091400757166,
      "grad_norm": 1.015625,
      "learning_rate": 0.000147,
      "loss": 7.0248,
      "mean_token_accuracy": 0.10310934484004974,
      "num_tokens": 665780.0,
      "step": 295
    },
    {
      "entropy": 7.552512168884277,
      "epoch": 0.32449972958355866,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0001495,
      "loss": 6.9293,
      "mean_token_accuracy": 0.10721782594919205,
      "num_tokens": 676566.0,
      "step": 300
    },
    {
      "entropy": 7.559945297241211,
      "epoch": 0.3299080584099513,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000152,
      "loss": 6.9251,
      "mean_token_accuracy": 0.1060689315199852,
      "num_tokens": 687407.0,
      "step": 305
    },
    {
      "entropy": 7.49097957611084,
      "epoch": 0.335316387236344,
      "grad_norm": 0.984375,
      "learning_rate": 0.00015450000000000001,
      "loss": 6.8497,
      "mean_token_accuracy": 0.11217133775353431,
      "num_tokens": 698162.0,
      "step": 310
    },
    {
      "entropy": 7.46426739692688,
      "epoch": 0.34072471606273663,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000157,
      "loss": 6.8316,
      "mean_token_accuracy": 0.10524874627590179,
      "num_tokens": 708743.0,
      "step": 315
    },
    {
      "entropy": 7.558863353729248,
      "epoch": 0.34613304488912927,
      "grad_norm": 1.09375,
      "learning_rate": 0.0001595,
      "loss": 7.0442,
      "mean_token_accuracy": 0.11182164400815964,
      "num_tokens": 719828.0,
      "step": 320
    },
    {
      "entropy": 7.583437824249268,
      "epoch": 0.3515413737155219,
      "grad_norm": 1.078125,
      "learning_rate": 0.000162,
      "loss": 6.9937,
      "mean_token_accuracy": 0.1039721980690956,
      "num_tokens": 732514.0,
      "step": 325
    },
    {
      "entropy": 7.449002027511597,
      "epoch": 0.35694970254191455,
      "grad_norm": 1.03125,
      "learning_rate": 0.00016450000000000001,
      "loss": 6.8317,
      "mean_token_accuracy": 0.11070109233260154,
      "num_tokens": 742391.0,
      "step": 330
    },
    {
      "entropy": 7.36940336227417,
      "epoch": 0.3623580313683072,
      "grad_norm": 1.1875,
      "learning_rate": 0.00016700000000000002,
      "loss": 6.8335,
      "mean_token_accuracy": 0.10628218725323677,
      "num_tokens": 753638.0,
      "step": 335
    },
    {
      "entropy": 7.5532920360565186,
      "epoch": 0.3677663601946998,
      "grad_norm": 1.109375,
      "learning_rate": 0.00016950000000000003,
      "loss": 7.0485,
      "mean_token_accuracy": 0.10574989169836044,
      "num_tokens": 764850.0,
      "step": 340
    },
    {
      "entropy": 7.4345468044281,
      "epoch": 0.37317468902109246,
      "grad_norm": 1.0625,
      "learning_rate": 0.00017199999999999998,
      "loss": 6.8115,
      "mean_token_accuracy": 0.11775769889354706,
      "num_tokens": 776192.0,
      "step": 345
    },
    {
      "entropy": 7.336619281768799,
      "epoch": 0.3785830178474851,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00017449999999999999,
      "loss": 6.7691,
      "mean_token_accuracy": 0.11614358723163605,
      "num_tokens": 787051.0,
      "step": 350
    },
    {
      "entropy": 7.326699829101562,
      "epoch": 0.3839913466738778,
      "grad_norm": 1.140625,
      "learning_rate": 0.000177,
      "loss": 6.7934,
      "mean_token_accuracy": 0.11272966936230659,
      "num_tokens": 799281.0,
      "step": 355
    },
    {
      "entropy": 7.331569719314575,
      "epoch": 0.38939967550027044,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0001795,
      "loss": 6.813,
      "mean_token_accuracy": 0.11107392236590385,
      "num_tokens": 810003.0,
      "step": 360
    },
    {
      "entropy": 7.336561441421509,
      "epoch": 0.3948080043266631,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000182,
      "loss": 6.7821,
      "mean_token_accuracy": 0.11116143241524697,
      "num_tokens": 819749.0,
      "step": 365
    },
    {
      "entropy": 7.329193019866944,
      "epoch": 0.4002163331530557,
      "grad_norm": 1.015625,
      "learning_rate": 0.0001845,
      "loss": 6.7888,
      "mean_token_accuracy": 0.11996297091245652,
      "num_tokens": 830552.0,
      "step": 370
    },
    {
      "entropy": 7.390970516204834,
      "epoch": 0.40562466197944835,
      "grad_norm": 1.3125,
      "learning_rate": 0.000187,
      "loss": 6.8901,
      "mean_token_accuracy": 0.1121357448399067,
      "num_tokens": 840495.0,
      "step": 375
    },
    {
      "entropy": 7.275109004974365,
      "epoch": 0.411032990805841,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0001895,
      "loss": 6.8114,
      "mean_token_accuracy": 0.1169951744377613,
      "num_tokens": 852915.0,
      "step": 380
    },
    {
      "entropy": 7.377492761611938,
      "epoch": 0.41644131963223363,
      "grad_norm": 1.203125,
      "learning_rate": 0.000192,
      "loss": 6.7822,
      "mean_token_accuracy": 0.11362324208021164,
      "num_tokens": 862762.0,
      "step": 385
    },
    {
      "entropy": 7.277709007263184,
      "epoch": 0.42184964845862627,
      "grad_norm": 1.171875,
      "learning_rate": 0.0001945,
      "loss": 6.7274,
      "mean_token_accuracy": 0.11747928410768509,
      "num_tokens": 873023.0,
      "step": 390
    },
    {
      "entropy": 7.28900933265686,
      "epoch": 0.4272579772850189,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00019700000000000002,
      "loss": 6.8274,
      "mean_token_accuracy": 0.10971275642514229,
      "num_tokens": 885281.0,
      "step": 395
    },
    {
      "entropy": 7.2772074222564695,
      "epoch": 0.43266630611141155,
      "grad_norm": 1.125,
      "learning_rate": 0.00019950000000000002,
      "loss": 6.8206,
      "mean_token_accuracy": 0.11502003371715545,
      "num_tokens": 896210.0,
      "step": 400
    },
    {
      "entropy": 7.189112377166748,
      "epoch": 0.43807463493780424,
      "grad_norm": 1.109375,
      "learning_rate": 0.000202,
      "loss": 6.7468,
      "mean_token_accuracy": 0.11514622867107391,
      "num_tokens": 909028.0,
      "step": 405
    },
    {
      "entropy": 7.229736566543579,
      "epoch": 0.4434829637641969,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00020449999999999998,
      "loss": 6.6815,
      "mean_token_accuracy": 0.11853564456105233,
      "num_tokens": 919710.0,
      "step": 410
    },
    {
      "entropy": 7.304251861572266,
      "epoch": 0.4488912925905895,
      "grad_norm": 1.09375,
      "learning_rate": 0.000207,
      "loss": 6.736,
      "mean_token_accuracy": 0.11397984251379967,
      "num_tokens": 930019.0,
      "step": 415
    },
    {
      "entropy": 7.166671371459961,
      "epoch": 0.45429962141698216,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0002095,
      "loss": 6.6524,
      "mean_token_accuracy": 0.11951824501156808,
      "num_tokens": 940535.0,
      "step": 420
    },
    {
      "entropy": 7.0795973777771,
      "epoch": 0.4597079502433748,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000212,
      "loss": 6.6119,
      "mean_token_accuracy": 0.12140549421310425,
      "num_tokens": 951390.0,
      "step": 425
    },
    {
      "entropy": 7.141013669967651,
      "epoch": 0.46511627906976744,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002145,
      "loss": 6.6636,
      "mean_token_accuracy": 0.1212614968419075,
      "num_tokens": 963010.0,
      "step": 430
    },
    {
      "entropy": 7.112875986099243,
      "epoch": 0.4705246078961601,
      "grad_norm": 1.0625,
      "learning_rate": 0.00021700000000000002,
      "loss": 6.6152,
      "mean_token_accuracy": 0.12599240839481354,
      "num_tokens": 974032.0,
      "step": 435
    },
    {
      "entropy": 7.222216367721558,
      "epoch": 0.4759329367225527,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002195,
      "loss": 6.6586,
      "mean_token_accuracy": 0.11954770833253861,
      "num_tokens": 983954.0,
      "step": 440
    },
    {
      "entropy": 7.11478705406189,
      "epoch": 0.48134126554894535,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000222,
      "loss": 6.689,
      "mean_token_accuracy": 0.12196131646633149,
      "num_tokens": 994135.0,
      "step": 445
    },
    {
      "entropy": 7.173604154586792,
      "epoch": 0.48674959437533805,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002245,
      "loss": 6.6493,
      "mean_token_accuracy": 0.12306504994630814,
      "num_tokens": 1006275.0,
      "step": 450
    },
    {
      "entropy": 6.988444137573242,
      "epoch": 0.4921579232017307,
      "grad_norm": 1.234375,
      "learning_rate": 0.00022700000000000002,
      "loss": 6.4801,
      "mean_token_accuracy": 0.129085473716259,
      "num_tokens": 1017767.0,
      "step": 455
    },
    {
      "entropy": 7.0636707782745365,
      "epoch": 0.4975662520281233,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00022950000000000002,
      "loss": 6.6038,
      "mean_token_accuracy": 0.1241903193295002,
      "num_tokens": 1028286.0,
      "step": 460
    },
    {
      "entropy": 7.059632682800293,
      "epoch": 0.502974580854516,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00023200000000000003,
      "loss": 6.6151,
      "mean_token_accuracy": 0.12029262557625771,
      "num_tokens": 1040555.0,
      "step": 465
    },
    {
      "entropy": 7.171442461013794,
      "epoch": 0.5083829096809086,
      "grad_norm": 0.984375,
      "learning_rate": 0.00023449999999999998,
      "loss": 6.7175,
      "mean_token_accuracy": 0.1209532730281353,
      "num_tokens": 1053668.0,
      "step": 470
    },
    {
      "entropy": 7.12706880569458,
      "epoch": 0.5137912385073012,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000237,
      "loss": 6.6597,
      "mean_token_accuracy": 0.11960283890366555,
      "num_tokens": 1066042.0,
      "step": 475
    },
    {
      "entropy": 6.984622383117676,
      "epoch": 0.5191995673336939,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002395,
      "loss": 6.6701,
      "mean_token_accuracy": 0.12059814557433128,
      "num_tokens": 1077843.0,
      "step": 480
    },
    {
      "entropy": 7.047416830062867,
      "epoch": 0.5246078961600865,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000242,
      "loss": 6.5605,
      "mean_token_accuracy": 0.1252925641834736,
      "num_tokens": 1089448.0,
      "step": 485
    },
    {
      "entropy": 7.041959238052368,
      "epoch": 0.5300162249864792,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002445,
      "loss": 6.6615,
      "mean_token_accuracy": 0.12150585278868675,
      "num_tokens": 1100989.0,
      "step": 490
    },
    {
      "entropy": 7.162778997421265,
      "epoch": 0.5354245538128718,
      "grad_norm": 1.046875,
      "learning_rate": 0.000247,
      "loss": 6.7004,
      "mean_token_accuracy": 0.1256784491240978,
      "num_tokens": 1113012.0,
      "step": 495
    },
    {
      "entropy": 7.056042337417603,
      "epoch": 0.5408328826392644,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002495,
      "loss": 6.6356,
      "mean_token_accuracy": 0.12231024205684662,
      "num_tokens": 1123551.0,
      "step": 500
    },
    {
      "epoch": 0.5408328826392644,
      "eval_entropy": 7.0043568221005525,
      "eval_loss": 6.780564308166504,
      "eval_mean_token_accuracy": 0.12075253836810589,
      "eval_num_tokens": 1123551.0,
      "eval_runtime": 1.9941,
      "eval_samples_per_second": 1762.721,
      "eval_steps_per_second": 220.654,
      "step": 500
    },
    {
      "entropy": 7.074879932403564,
      "epoch": 0.5462412114656571,
      "grad_norm": 1.359375,
      "learning_rate": 0.000252,
      "loss": 6.5659,
      "mean_token_accuracy": 0.12208069637417793,
      "num_tokens": 1133838.0,
      "step": 505
    },
    {
      "entropy": 6.973394107818604,
      "epoch": 0.5516495402920497,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0002545,
      "loss": 6.6027,
      "mean_token_accuracy": 0.12259892076253891,
      "num_tokens": 1146005.0,
      "step": 510
    },
    {
      "entropy": 7.00295934677124,
      "epoch": 0.5570578691184424,
      "grad_norm": 1.125,
      "learning_rate": 0.000257,
      "loss": 6.6885,
      "mean_token_accuracy": 0.1244613103568554,
      "num_tokens": 1157328.0,
      "step": 515
    },
    {
      "entropy": 7.057001447677612,
      "epoch": 0.5624661979448351,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0002595,
      "loss": 6.4842,
      "mean_token_accuracy": 0.1322474293410778,
      "num_tokens": 1168592.0,
      "step": 520
    },
    {
      "entropy": 7.032931184768676,
      "epoch": 0.5678745267712277,
      "grad_norm": 1.125,
      "learning_rate": 0.000262,
      "loss": 6.6161,
      "mean_token_accuracy": 0.12481116801500321,
      "num_tokens": 1178680.0,
      "step": 525
    },
    {
      "entropy": 6.9781468391418455,
      "epoch": 0.5732828555976204,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00026450000000000003,
      "loss": 6.5535,
      "mean_token_accuracy": 0.127150696516037,
      "num_tokens": 1188825.0,
      "step": 530
    },
    {
      "entropy": 7.008031415939331,
      "epoch": 0.578691184424013,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00026700000000000004,
      "loss": 6.5956,
      "mean_token_accuracy": 0.12398400753736497,
      "num_tokens": 1201375.0,
      "step": 535
    },
    {
      "entropy": 7.072990751266479,
      "epoch": 0.5840995132504057,
      "grad_norm": 1.125,
      "learning_rate": 0.00026950000000000005,
      "loss": 6.7227,
      "mean_token_accuracy": 0.11844412088394166,
      "num_tokens": 1211915.0,
      "step": 540
    },
    {
      "entropy": 6.957782316207886,
      "epoch": 0.5895078420767983,
      "grad_norm": 1.21875,
      "learning_rate": 0.00027200000000000005,
      "loss": 6.6289,
      "mean_token_accuracy": 0.12811450883746148,
      "num_tokens": 1223151.0,
      "step": 545
    },
    {
      "entropy": 7.041991472244263,
      "epoch": 0.5949161709031909,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002745,
      "loss": 6.5414,
      "mean_token_accuracy": 0.13184184953570366,
      "num_tokens": 1234229.0,
      "step": 550
    },
    {
      "entropy": 7.023768568038941,
      "epoch": 0.6003244997295836,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000277,
      "loss": 6.5529,
      "mean_token_accuracy": 0.13157471716403962,
      "num_tokens": 1245298.0,
      "step": 555
    },
    {
      "entropy": 6.8409104347229,
      "epoch": 0.6057328285559762,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002795,
      "loss": 6.5595,
      "mean_token_accuracy": 0.12466495037078858,
      "num_tokens": 1256154.0,
      "step": 560
    },
    {
      "entropy": 6.922661924362183,
      "epoch": 0.6111411573823688,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00028199999999999997,
      "loss": 6.4705,
      "mean_token_accuracy": 0.129321701079607,
      "num_tokens": 1267404.0,
      "step": 565
    },
    {
      "entropy": 6.944332456588745,
      "epoch": 0.6165494862087615,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002845,
      "loss": 6.5185,
      "mean_token_accuracy": 0.12794106900691987,
      "num_tokens": 1278894.0,
      "step": 570
    },
    {
      "entropy": 6.9090385913848875,
      "epoch": 0.6219578150351541,
      "grad_norm": 1.125,
      "learning_rate": 0.000287,
      "loss": 6.502,
      "mean_token_accuracy": 0.13496915996074677,
      "num_tokens": 1290408.0,
      "step": 575
    },
    {
      "entropy": 6.971935558319092,
      "epoch": 0.6273661438615468,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002895,
      "loss": 6.5806,
      "mean_token_accuracy": 0.12681553587317468,
      "num_tokens": 1301508.0,
      "step": 580
    },
    {
      "entropy": 6.989264011383057,
      "epoch": 0.6327744726879394,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000292,
      "loss": 6.5447,
      "mean_token_accuracy": 0.13189456537365912,
      "num_tokens": 1312551.0,
      "step": 585
    },
    {
      "entropy": 6.965842962265015,
      "epoch": 0.638182801514332,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002945,
      "loss": 6.6459,
      "mean_token_accuracy": 0.12647571191191673,
      "num_tokens": 1324876.0,
      "step": 590
    },
    {
      "entropy": 6.939603042602539,
      "epoch": 0.6435911303407247,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000297,
      "loss": 6.531,
      "mean_token_accuracy": 0.12691377103328705,
      "num_tokens": 1335401.0,
      "step": 595
    },
    {
      "entropy": 6.8328300476074215,
      "epoch": 0.6489994591671173,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002995,
      "loss": 6.4707,
      "mean_token_accuracy": 0.127620816975832,
      "num_tokens": 1348265.0,
      "step": 600
    },
    {
      "entropy": 6.7745624542236325,
      "epoch": 0.65440778799351,
      "grad_norm": 1.109375,
      "learning_rate": 0.000302,
      "loss": 6.4306,
      "mean_token_accuracy": 0.13703566640615464,
      "num_tokens": 1359078.0,
      "step": 605
    },
    {
      "entropy": 6.712151575088501,
      "epoch": 0.6598161168199026,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003045,
      "loss": 6.3318,
      "mean_token_accuracy": 0.13108599931001663,
      "num_tokens": 1369499.0,
      "step": 610
    },
    {
      "entropy": 6.8127624034881595,
      "epoch": 0.6652244456462953,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000307,
      "loss": 6.4323,
      "mean_token_accuracy": 0.1286877028644085,
      "num_tokens": 1380329.0,
      "step": 615
    },
    {
      "entropy": 6.824408864974975,
      "epoch": 0.670632774472688,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003095,
      "loss": 6.5116,
      "mean_token_accuracy": 0.12648341283202172,
      "num_tokens": 1393047.0,
      "step": 620
    },
    {
      "entropy": 6.872823572158813,
      "epoch": 0.6760411032990806,
      "grad_norm": 0.9140625,
      "learning_rate": 0.000312,
      "loss": 6.5067,
      "mean_token_accuracy": 0.12922688201069832,
      "num_tokens": 1405600.0,
      "step": 625
    },
    {
      "entropy": 6.811132955551147,
      "epoch": 0.6814494321254733,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003145,
      "loss": 6.4417,
      "mean_token_accuracy": 0.13407272696495057,
      "num_tokens": 1416670.0,
      "step": 630
    },
    {
      "entropy": 6.825539922714233,
      "epoch": 0.6868577609518659,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000317,
      "loss": 6.4599,
      "mean_token_accuracy": 0.12846649512648584,
      "num_tokens": 1428929.0,
      "step": 635
    },
    {
      "entropy": 6.827125072479248,
      "epoch": 0.6922660897782585,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003195,
      "loss": 6.3887,
      "mean_token_accuracy": 0.13251324519515037,
      "num_tokens": 1439298.0,
      "step": 640
    },
    {
      "entropy": 6.817316055297852,
      "epoch": 0.6976744186046512,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000322,
      "loss": 6.4844,
      "mean_token_accuracy": 0.13544763550162314,
      "num_tokens": 1449853.0,
      "step": 645
    },
    {
      "entropy": 6.891280555725098,
      "epoch": 0.7030827474310438,
      "grad_norm": 1.171875,
      "learning_rate": 0.00032450000000000003,
      "loss": 6.5515,
      "mean_token_accuracy": 0.12688857838511466,
      "num_tokens": 1460816.0,
      "step": 650
    },
    {
      "entropy": 6.740885925292969,
      "epoch": 0.7084910762574365,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00032700000000000003,
      "loss": 6.4014,
      "mean_token_accuracy": 0.13262718617916108,
      "num_tokens": 1473003.0,
      "step": 655
    },
    {
      "entropy": 6.843165111541748,
      "epoch": 0.7138994050838291,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00032950000000000004,
      "loss": 6.5206,
      "mean_token_accuracy": 0.1229576200246811,
      "num_tokens": 1484745.0,
      "step": 660
    },
    {
      "entropy": 6.823391914367676,
      "epoch": 0.7193077339102217,
      "grad_norm": 1.03125,
      "learning_rate": 0.00033200000000000005,
      "loss": 6.4011,
      "mean_token_accuracy": 0.13305559009313583,
      "num_tokens": 1495803.0,
      "step": 665
    },
    {
      "entropy": 6.803869295120239,
      "epoch": 0.7247160627366144,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00033450000000000005,
      "loss": 6.4092,
      "mean_token_accuracy": 0.141731408983469,
      "num_tokens": 1507504.0,
      "step": 670
    },
    {
      "entropy": 6.852431774139404,
      "epoch": 0.730124391563007,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000337,
      "loss": 6.5033,
      "mean_token_accuracy": 0.12678006961941718,
      "num_tokens": 1518765.0,
      "step": 675
    },
    {
      "entropy": 6.650970602035523,
      "epoch": 0.7355327203893997,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003395,
      "loss": 6.3839,
      "mean_token_accuracy": 0.1324382781982422,
      "num_tokens": 1529481.0,
      "step": 680
    },
    {
      "entropy": 6.810516691207885,
      "epoch": 0.7409410492157923,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000342,
      "loss": 6.3986,
      "mean_token_accuracy": 0.13725290223956108,
      "num_tokens": 1539612.0,
      "step": 685
    },
    {
      "entropy": 6.801828384399414,
      "epoch": 0.7463493780421849,
      "grad_norm": 0.984375,
      "learning_rate": 0.00034449999999999997,
      "loss": 6.558,
      "mean_token_accuracy": 0.12963267266750336,
      "num_tokens": 1551370.0,
      "step": 690
    },
    {
      "entropy": 6.8387431621551515,
      "epoch": 0.7517577068685776,
      "grad_norm": 1.0625,
      "learning_rate": 0.000347,
      "loss": 6.44,
      "mean_token_accuracy": 0.1378513492643833,
      "num_tokens": 1561662.0,
      "step": 695
    },
    {
      "entropy": 6.7489661693573,
      "epoch": 0.7571660356949702,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0003495,
      "loss": 6.4232,
      "mean_token_accuracy": 0.13230135440826415,
      "num_tokens": 1573479.0,
      "step": 700
    },
    {
      "entropy": 6.756485414505005,
      "epoch": 0.7625743645213628,
      "grad_norm": 1.203125,
      "learning_rate": 0.000352,
      "loss": 6.366,
      "mean_token_accuracy": 0.1389087036252022,
      "num_tokens": 1583684.0,
      "step": 705
    },
    {
      "entropy": 6.7295139789581295,
      "epoch": 0.7679826933477556,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003545,
      "loss": 6.4101,
      "mean_token_accuracy": 0.13583654090762137,
      "num_tokens": 1595694.0,
      "step": 710
    },
    {
      "entropy": 6.696416282653809,
      "epoch": 0.7733910221741482,
      "grad_norm": 1.25,
      "learning_rate": 0.000357,
      "loss": 6.4184,
      "mean_token_accuracy": 0.13240900188684462,
      "num_tokens": 1606104.0,
      "step": 715
    },
    {
      "entropy": 6.7834718227386475,
      "epoch": 0.7787993510005409,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003595,
      "loss": 6.4803,
      "mean_token_accuracy": 0.13445528894662856,
      "num_tokens": 1617368.0,
      "step": 720
    },
    {
      "entropy": 6.790578842163086,
      "epoch": 0.7842076798269335,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000362,
      "loss": 6.4893,
      "mean_token_accuracy": 0.1305526874959469,
      "num_tokens": 1630485.0,
      "step": 725
    },
    {
      "entropy": 6.788717746734619,
      "epoch": 0.7896160086533262,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003645,
      "loss": 6.3357,
      "mean_token_accuracy": 0.1369060456752777,
      "num_tokens": 1641411.0,
      "step": 730
    },
    {
      "entropy": 6.644323205947876,
      "epoch": 0.7950243374797188,
      "grad_norm": 1.078125,
      "learning_rate": 0.000367,
      "loss": 6.3465,
      "mean_token_accuracy": 0.13824027627706528,
      "num_tokens": 1653737.0,
      "step": 735
    },
    {
      "entropy": 6.823546600341797,
      "epoch": 0.8004326663061114,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003695,
      "loss": 6.5281,
      "mean_token_accuracy": 0.12467064261436463,
      "num_tokens": 1665910.0,
      "step": 740
    },
    {
      "entropy": 6.726630735397339,
      "epoch": 0.8058409951325041,
      "grad_norm": 1.109375,
      "learning_rate": 0.000372,
      "loss": 6.3094,
      "mean_token_accuracy": 0.134908963739872,
      "num_tokens": 1676927.0,
      "step": 745
    },
    {
      "entropy": 6.629592800140381,
      "epoch": 0.8112493239588967,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003745,
      "loss": 6.357,
      "mean_token_accuracy": 0.13599003627896308,
      "num_tokens": 1688167.0,
      "step": 750
    },
    {
      "entropy": 6.650397443771363,
      "epoch": 0.8166576527852893,
      "grad_norm": 1.0,
      "learning_rate": 0.000377,
      "loss": 6.4308,
      "mean_token_accuracy": 0.13610320165753365,
      "num_tokens": 1699899.0,
      "step": 755
    },
    {
      "entropy": 6.7162903308868405,
      "epoch": 0.822065981611682,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003795,
      "loss": 6.3533,
      "mean_token_accuracy": 0.1412688173353672,
      "num_tokens": 1709862.0,
      "step": 760
    },
    {
      "entropy": 6.640978765487671,
      "epoch": 0.8274743104380746,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000382,
      "loss": 6.3481,
      "mean_token_accuracy": 0.1376900367438793,
      "num_tokens": 1719902.0,
      "step": 765
    },
    {
      "entropy": 6.691372346878052,
      "epoch": 0.8328826392644673,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003845,
      "loss": 6.3541,
      "mean_token_accuracy": 0.13510108962655068,
      "num_tokens": 1730351.0,
      "step": 770
    },
    {
      "entropy": 6.589955043792725,
      "epoch": 0.8382909680908599,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00038700000000000003,
      "loss": 6.3233,
      "mean_token_accuracy": 0.1341069035232067,
      "num_tokens": 1741783.0,
      "step": 775
    },
    {
      "entropy": 6.609476327896118,
      "epoch": 0.8436992969172525,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00038950000000000003,
      "loss": 6.3639,
      "mean_token_accuracy": 0.13918454721570014,
      "num_tokens": 1752952.0,
      "step": 780
    },
    {
      "entropy": 6.713692474365234,
      "epoch": 0.8491076257436452,
      "grad_norm": 1.125,
      "learning_rate": 0.00039200000000000004,
      "loss": 6.3796,
      "mean_token_accuracy": 0.1377579912543297,
      "num_tokens": 1764014.0,
      "step": 785
    },
    {
      "entropy": 6.692055416107178,
      "epoch": 0.8545159545700378,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00039450000000000005,
      "loss": 6.3501,
      "mean_token_accuracy": 0.13647191673517228,
      "num_tokens": 1776154.0,
      "step": 790
    },
    {
      "entropy": 6.651118326187134,
      "epoch": 0.8599242833964305,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00039700000000000005,
      "loss": 6.405,
      "mean_token_accuracy": 0.1311624176800251,
      "num_tokens": 1787292.0,
      "step": 795
    },
    {
      "entropy": 6.63629379272461,
      "epoch": 0.8653326122228231,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003995,
      "loss": 6.3667,
      "mean_token_accuracy": 0.13610742837190629,
      "num_tokens": 1799469.0,
      "step": 800
    },
    {
      "entropy": 6.567807674407959,
      "epoch": 0.8707409410492158,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000402,
      "loss": 6.2441,
      "mean_token_accuracy": 0.14459661692380904,
      "num_tokens": 1810091.0,
      "step": 805
    },
    {
      "entropy": 6.723603248596191,
      "epoch": 0.8761492698756085,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004045,
      "loss": 6.3925,
      "mean_token_accuracy": 0.14025203064084052,
      "num_tokens": 1820835.0,
      "step": 810
    },
    {
      "entropy": 6.633598756790161,
      "epoch": 0.8815575987020011,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040699999999999997,
      "loss": 6.3549,
      "mean_token_accuracy": 0.13814282566308975,
      "num_tokens": 1831868.0,
      "step": 815
    },
    {
      "entropy": 6.587992763519287,
      "epoch": 0.8869659275283938,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004095,
      "loss": 6.3383,
      "mean_token_accuracy": 0.13848361223936081,
      "num_tokens": 1842040.0,
      "step": 820
    },
    {
      "entropy": 6.723627090454102,
      "epoch": 0.8923742563547864,
      "grad_norm": 1.015625,
      "learning_rate": 0.000412,
      "loss": 6.4083,
      "mean_token_accuracy": 0.13382988572120666,
      "num_tokens": 1853765.0,
      "step": 825
    },
    {
      "entropy": 6.673075199127197,
      "epoch": 0.897782585181179,
      "grad_norm": 1.0,
      "learning_rate": 0.0004145,
      "loss": 6.4141,
      "mean_token_accuracy": 0.13732750713825226,
      "num_tokens": 1866081.0,
      "step": 830
    },
    {
      "entropy": 6.593461990356445,
      "epoch": 0.9031909140075717,
      "grad_norm": 1.125,
      "learning_rate": 0.000417,
      "loss": 6.3155,
      "mean_token_accuracy": 0.13803164809942245,
      "num_tokens": 1876754.0,
      "step": 835
    },
    {
      "entropy": 6.714284658432007,
      "epoch": 0.9085992428339643,
      "grad_norm": 1.125,
      "learning_rate": 0.0004195,
      "loss": 6.4263,
      "mean_token_accuracy": 0.1334732621908188,
      "num_tokens": 1888299.0,
      "step": 840
    },
    {
      "entropy": 6.475223159790039,
      "epoch": 0.914007571660357,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000422,
      "loss": 6.2122,
      "mean_token_accuracy": 0.1465678930282593,
      "num_tokens": 1899027.0,
      "step": 845
    },
    {
      "entropy": 6.625812816619873,
      "epoch": 0.9194159004867496,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004245,
      "loss": 6.3374,
      "mean_token_accuracy": 0.14080506712198257,
      "num_tokens": 1911043.0,
      "step": 850
    },
    {
      "entropy": 6.590130615234375,
      "epoch": 0.9248242293131422,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000427,
      "loss": 6.2567,
      "mean_token_accuracy": 0.13707533478736877,
      "num_tokens": 1921295.0,
      "step": 855
    },
    {
      "entropy": 6.61051025390625,
      "epoch": 0.9302325581395349,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004295,
      "loss": 6.3569,
      "mean_token_accuracy": 0.13993956744670868,
      "num_tokens": 1932944.0,
      "step": 860
    },
    {
      "entropy": 6.553638172149658,
      "epoch": 0.9356408869659275,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000432,
      "loss": 6.3027,
      "mean_token_accuracy": 0.13260096460580825,
      "num_tokens": 1944072.0,
      "step": 865
    },
    {
      "entropy": 6.569744491577149,
      "epoch": 0.9410492157923201,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004345,
      "loss": 6.2956,
      "mean_token_accuracy": 0.13898471891880035,
      "num_tokens": 1954798.0,
      "step": 870
    },
    {
      "entropy": 6.513964128494263,
      "epoch": 0.9464575446187128,
      "grad_norm": 1.03125,
      "learning_rate": 0.000437,
      "loss": 6.2698,
      "mean_token_accuracy": 0.1473902389407158,
      "num_tokens": 1965304.0,
      "step": 875
    },
    {
      "entropy": 6.6614869117736815,
      "epoch": 0.9518658734451054,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004395,
      "loss": 6.3094,
      "mean_token_accuracy": 0.13569752722978592,
      "num_tokens": 1976797.0,
      "step": 880
    },
    {
      "entropy": 6.550220966339111,
      "epoch": 0.9572742022714981,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000442,
      "loss": 6.3248,
      "mean_token_accuracy": 0.13910794109106064,
      "num_tokens": 1987448.0,
      "step": 885
    },
    {
      "entropy": 6.369886207580566,
      "epoch": 0.9626825310978907,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004445,
      "loss": 6.1508,
      "mean_token_accuracy": 0.1404670625925064,
      "num_tokens": 2000948.0,
      "step": 890
    },
    {
      "entropy": 6.55863881111145,
      "epoch": 0.9680908599242833,
      "grad_norm": 1.046875,
      "learning_rate": 0.000447,
      "loss": 6.2643,
      "mean_token_accuracy": 0.14369212165474893,
      "num_tokens": 2011807.0,
      "step": 895
    },
    {
      "entropy": 6.6163318157196045,
      "epoch": 0.9734991887506761,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044950000000000003,
      "loss": 6.3207,
      "mean_token_accuracy": 0.13593853637576103,
      "num_tokens": 2022835.0,
      "step": 900
    },
    {
      "entropy": 6.3906701564788815,
      "epoch": 0.9789075175770687,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00045200000000000004,
      "loss": 6.2339,
      "mean_token_accuracy": 0.14182498455047607,
      "num_tokens": 2034331.0,
      "step": 905
    },
    {
      "entropy": 6.662575387954712,
      "epoch": 0.9843158464034614,
      "grad_norm": 1.0,
      "learning_rate": 0.00045450000000000004,
      "loss": 6.2328,
      "mean_token_accuracy": 0.13782825246453284,
      "num_tokens": 2047002.0,
      "step": 910
    },
    {
      "entropy": 6.408291864395141,
      "epoch": 0.989724175229854,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045700000000000005,
      "loss": 6.1835,
      "mean_token_accuracy": 0.14051275625824927,
      "num_tokens": 2056299.0,
      "step": 915
    },
    {
      "entropy": 6.519381284713745,
      "epoch": 0.9951325040562466,
      "grad_norm": 1.0,
      "learning_rate": 0.00045950000000000006,
      "loss": 6.3108,
      "mean_token_accuracy": 0.13649527579545975,
      "num_tokens": 2068031.0,
      "step": 920
    },
    {
      "entropy": 6.533435503641765,
      "epoch": 1.0,
      "grad_norm": 1.8125,
      "learning_rate": 0.000462,
      "loss": 6.2559,
      "mean_token_accuracy": 0.1411174734433492,
      "num_tokens": 2076889.0,
      "step": 925
    },
    {
      "entropy": 6.518666791915893,
      "epoch": 1.0054083288263926,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004645,
      "loss": 6.1187,
      "mean_token_accuracy": 0.14975121915340422,
      "num_tokens": 2087316.0,
      "step": 930
    },
    {
      "entropy": 6.499506664276123,
      "epoch": 1.0108166576527853,
      "grad_norm": 1.03125,
      "learning_rate": 0.000467,
      "loss": 6.1223,
      "mean_token_accuracy": 0.14297691583633423,
      "num_tokens": 2098522.0,
      "step": 935
    },
    {
      "entropy": 6.383229446411133,
      "epoch": 1.016224986479178,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004695,
      "loss": 6.0878,
      "mean_token_accuracy": 0.15435394942760466,
      "num_tokens": 2110015.0,
      "step": 940
    },
    {
      "entropy": 6.525584554672241,
      "epoch": 1.0216333153055706,
      "grad_norm": 1.0625,
      "learning_rate": 0.000472,
      "loss": 6.1924,
      "mean_token_accuracy": 0.1424684338271618,
      "num_tokens": 2121333.0,
      "step": 945
    },
    {
      "entropy": 6.4724767208099365,
      "epoch": 1.0270416441319632,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004745,
      "loss": 6.0838,
      "mean_token_accuracy": 0.14994525909423828,
      "num_tokens": 2131385.0,
      "step": 950
    },
    {
      "entropy": 6.488972854614258,
      "epoch": 1.0324499729583558,
      "grad_norm": 1.015625,
      "learning_rate": 0.000477,
      "loss": 6.1275,
      "mean_token_accuracy": 0.1522180952131748,
      "num_tokens": 2143274.0,
      "step": 955
    },
    {
      "entropy": 6.368765354156494,
      "epoch": 1.0378583017847485,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004795,
      "loss": 6.0124,
      "mean_token_accuracy": 0.15076057612895966,
      "num_tokens": 2154529.0,
      "step": 960
    },
    {
      "entropy": 6.4038872718811035,
      "epoch": 1.043266630611141,
      "grad_norm": 1.078125,
      "learning_rate": 0.000482,
      "loss": 6.1603,
      "mean_token_accuracy": 0.1439424842596054,
      "num_tokens": 2165949.0,
      "step": 965
    },
    {
      "entropy": 6.407582426071167,
      "epoch": 1.0486749594375337,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004845,
      "loss": 6.1335,
      "mean_token_accuracy": 0.1485764428973198,
      "num_tokens": 2176710.0,
      "step": 970
    },
    {
      "entropy": 6.463790941238403,
      "epoch": 1.0540832882639264,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000487,
      "loss": 6.1074,
      "mean_token_accuracy": 0.15412099212408065,
      "num_tokens": 2189470.0,
      "step": 975
    },
    {
      "entropy": 6.383792352676392,
      "epoch": 1.059491617090319,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004895,
      "loss": 6.0844,
      "mean_token_accuracy": 0.1579994946718216,
      "num_tokens": 2200266.0,
      "step": 980
    },
    {
      "entropy": 6.395930194854737,
      "epoch": 1.0648999459167117,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000492,
      "loss": 6.0915,
      "mean_token_accuracy": 0.14566799476742745,
      "num_tokens": 2210736.0,
      "step": 985
    },
    {
      "entropy": 6.478207111358643,
      "epoch": 1.0703082747431043,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004945,
      "loss": 6.1985,
      "mean_token_accuracy": 0.14791767448186874,
      "num_tokens": 2221057.0,
      "step": 990
    },
    {
      "entropy": 6.455506134033203,
      "epoch": 1.075716603569497,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000497,
      "loss": 6.1738,
      "mean_token_accuracy": 0.14685477688908577,
      "num_tokens": 2232352.0,
      "step": 995
    },
    {
      "entropy": 6.510460042953492,
      "epoch": 1.0811249323958896,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004995,
      "loss": 6.0805,
      "mean_token_accuracy": 0.15254651606082917,
      "num_tokens": 2244655.0,
      "step": 1000
    },
    {
      "epoch": 1.0811249323958896,
      "eval_entropy": 6.124205052852631,
      "eval_loss": 6.387869834899902,
      "eval_mean_token_accuracy": 0.1444568099030717,
      "eval_num_tokens": 2244655.0,
      "eval_runtime": 1.9538,
      "eval_samples_per_second": 1799.022,
      "eval_steps_per_second": 225.198,
      "step": 1000
    },
    {
      "entropy": 6.2928147315979,
      "epoch": 1.0865332612222822,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999997383518051,
      "loss": 6.0382,
      "mean_token_accuracy": 0.1600428655743599,
      "num_tokens": 2255699.0,
      "step": 1005
    },
    {
      "entropy": 6.372293472290039,
      "epoch": 1.0919415900486749,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999986754070562,
      "loss": 6.0381,
      "mean_token_accuracy": 0.15901420414447784,
      "num_tokens": 2266627.0,
      "step": 1010
    },
    {
      "entropy": 6.441066265106201,
      "epoch": 1.0973499188750675,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004999967948166009,
      "loss": 6.2142,
      "mean_token_accuracy": 0.14456040039658546,
      "num_tokens": 2277814.0,
      "step": 1015
    },
    {
      "entropy": 6.525017118453979,
      "epoch": 1.1027582477014604,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999940965872734,
      "loss": 6.1616,
      "mean_token_accuracy": 0.14813328310847282,
      "num_tokens": 2287344.0,
      "step": 1020
    },
    {
      "entropy": 6.212237882614136,
      "epoch": 1.1081665765278528,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999905807288788,
      "loss": 5.9498,
      "mean_token_accuracy": 0.1576882466673851,
      "num_tokens": 2298332.0,
      "step": 1025
    },
    {
      "entropy": 6.400769948959351,
      "epoch": 1.1135749053542456,
      "grad_norm": 1.03125,
      "learning_rate": 0.000499986247254194,
      "loss": 6.0778,
      "mean_token_accuracy": 0.15378758013248445,
      "num_tokens": 2309161.0,
      "step": 1030
    },
    {
      "entropy": 6.346611022949219,
      "epoch": 1.1189832341806383,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004999810961789667,
      "loss": 6.0663,
      "mean_token_accuracy": 0.15356869399547576,
      "num_tokens": 2319516.0,
      "step": 1035
    },
    {
      "entropy": 6.348343276977539,
      "epoch": 1.124391563007031,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000499975127521916,
      "loss": 5.9687,
      "mean_token_accuracy": 0.15849437564611435,
      "num_tokens": 2330441.0,
      "step": 1040
    },
    {
      "entropy": 6.287055778503418,
      "epoch": 1.1297998918334236,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999683413047319,
      "loss": 5.9837,
      "mean_token_accuracy": 0.15973089039325714,
      "num_tokens": 2340524.0,
      "step": 1045
    },
    {
      "entropy": 6.308336925506592,
      "epoch": 1.1352082206598162,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004999607375520759,
      "loss": 5.981,
      "mean_token_accuracy": 0.1552950382232666,
      "num_tokens": 2353379.0,
      "step": 1050
    },
    {
      "entropy": 6.3794340133667,
      "epoch": 1.1406165494862088,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999523162915794,
      "loss": 6.1602,
      "mean_token_accuracy": 0.1466098792850971,
      "num_tokens": 2365708.0,
      "step": 1055
    },
    {
      "entropy": 6.301113653182983,
      "epoch": 1.1460248783126015,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999430775538459,
      "loss": 6.0682,
      "mean_token_accuracy": 0.1489157885313034,
      "num_tokens": 2375630.0,
      "step": 1060
    },
    {
      "entropy": 6.2834553718566895,
      "epoch": 1.151433207138994,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999330213724487,
      "loss": 6.0778,
      "mean_token_accuracy": 0.14907022714614868,
      "num_tokens": 2386885.0,
      "step": 1065
    },
    {
      "entropy": 6.3159809589385985,
      "epoch": 1.1568415359653867,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999221477839319,
      "loss": 6.0987,
      "mean_token_accuracy": 0.1497051790356636,
      "num_tokens": 2397676.0,
      "step": 1070
    },
    {
      "entropy": 6.429526996612549,
      "epoch": 1.1622498647917794,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004999104568278104,
      "loss": 6.1048,
      "mean_token_accuracy": 0.14880598485469818,
      "num_tokens": 2409039.0,
      "step": 1075
    },
    {
      "entropy": 6.388683223724366,
      "epoch": 1.167658193618172,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000499897948546569,
      "loss": 6.1315,
      "mean_token_accuracy": 0.15224068611860275,
      "num_tokens": 2420103.0,
      "step": 1080
    },
    {
      "entropy": 6.3003730297088625,
      "epoch": 1.1730665224445647,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004998846229856629,
      "loss": 6.0491,
      "mean_token_accuracy": 0.15122028589248657,
      "num_tokens": 2430792.0,
      "step": 1085
    },
    {
      "entropy": 6.4121685981750485,
      "epoch": 1.1784748512709573,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004998704801935171,
      "loss": 6.0602,
      "mean_token_accuracy": 0.15023664385080338,
      "num_tokens": 2442165.0,
      "step": 1090
    },
    {
      "entropy": 6.261796760559082,
      "epoch": 1.18388318009735,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004998555202215267,
      "loss": 5.9712,
      "mean_token_accuracy": 0.15477441847324372,
      "num_tokens": 2454231.0,
      "step": 1095
    },
    {
      "entropy": 6.338596963882447,
      "epoch": 1.1892915089237426,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004998397431240561,
      "loss": 6.0251,
      "mean_token_accuracy": 0.15698925256729127,
      "num_tokens": 2464828.0,
      "step": 1100
    },
    {
      "entropy": 6.240490484237671,
      "epoch": 1.1946998377501352,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004998231489584395,
      "loss": 6.0689,
      "mean_token_accuracy": 0.14833179712295533,
      "num_tokens": 2476144.0,
      "step": 1105
    },
    {
      "entropy": 6.455478239059448,
      "epoch": 1.2001081665765279,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00049980573778498,
      "loss": 6.2395,
      "mean_token_accuracy": 0.1458302244544029,
      "num_tokens": 2489116.0,
      "step": 1110
    },
    {
      "entropy": 6.402680921554565,
      "epoch": 1.2055164954029205,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004997875096669498,
      "loss": 6.1098,
      "mean_token_accuracy": 0.15058583766222,
      "num_tokens": 2501609.0,
      "step": 1115
    },
    {
      "entropy": 6.230094385147095,
      "epoch": 1.2109248242293131,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004997684646705901,
      "loss": 6.0114,
      "mean_token_accuracy": 0.15897049903869628,
      "num_tokens": 2513338.0,
      "step": 1120
    },
    {
      "entropy": 6.354293441772461,
      "epoch": 1.2163331530557058,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004997486028651103,
      "loss": 6.0383,
      "mean_token_accuracy": 0.16039613038301467,
      "num_tokens": 2523199.0,
      "step": 1125
    },
    {
      "entropy": 6.349422979354858,
      "epoch": 1.2217414818820984,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004997279243226884,
      "loss": 6.0512,
      "mean_token_accuracy": 0.15354744642972945,
      "num_tokens": 2535136.0,
      "step": 1130
    },
    {
      "entropy": 6.264004421234131,
      "epoch": 1.227149810708491,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004997064291184701,
      "loss": 5.9657,
      "mean_token_accuracy": 0.14787945449352263,
      "num_tokens": 2546428.0,
      "step": 1135
    },
    {
      "entropy": 6.358812093734741,
      "epoch": 1.2325581395348837,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004996841173305691,
      "loss": 6.1636,
      "mean_token_accuracy": 0.15003497749567032,
      "num_tokens": 2558304.0,
      "step": 1140
    },
    {
      "entropy": 6.174890947341919,
      "epoch": 1.2379664683612763,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004996609890400665,
      "loss": 5.848,
      "mean_token_accuracy": 0.16567099541425706,
      "num_tokens": 2568717.0,
      "step": 1145
    },
    {
      "entropy": 6.297175502777099,
      "epoch": 1.243374797187669,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004996370443310107,
      "loss": 6.0533,
      "mean_token_accuracy": 0.14965572953224182,
      "num_tokens": 2578748.0,
      "step": 1150
    },
    {
      "entropy": 6.327052450180053,
      "epoch": 1.2487831260140616,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004996122832904165,
      "loss": 6.0212,
      "mean_token_accuracy": 0.1520503729581833,
      "num_tokens": 2590095.0,
      "step": 1155
    },
    {
      "entropy": 6.300268507003784,
      "epoch": 1.2541914548404542,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004995867060082658,
      "loss": 6.0529,
      "mean_token_accuracy": 0.15308669656515123,
      "num_tokens": 2600929.0,
      "step": 1160
    },
    {
      "entropy": 6.392643833160401,
      "epoch": 1.2595997836668469,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004995603125775065,
      "loss": 6.0596,
      "mean_token_accuracy": 0.15196561813354492,
      "num_tokens": 2611830.0,
      "step": 1165
    },
    {
      "entropy": 6.176296854019165,
      "epoch": 1.2650081124932395,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004995331030940523,
      "loss": 6.0102,
      "mean_token_accuracy": 0.1498885288834572,
      "num_tokens": 2623177.0,
      "step": 1170
    },
    {
      "entropy": 6.219478797912598,
      "epoch": 1.2704164413196322,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004995050776567826,
      "loss": 6.0001,
      "mean_token_accuracy": 0.15188704282045365,
      "num_tokens": 2635563.0,
      "step": 1175
    },
    {
      "entropy": 6.3707170486450195,
      "epoch": 1.2758247701460248,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004994762363675421,
      "loss": 5.9911,
      "mean_token_accuracy": 0.15853488594293594,
      "num_tokens": 2647157.0,
      "step": 1180
    },
    {
      "entropy": 6.177279806137085,
      "epoch": 1.2812330989724177,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004994465793311396,
      "loss": 5.9629,
      "mean_token_accuracy": 0.16011573225259781,
      "num_tokens": 2658175.0,
      "step": 1185
    },
    {
      "entropy": 6.249725675582885,
      "epoch": 1.28664142779881,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004994161066553493,
      "loss": 5.9191,
      "mean_token_accuracy": 0.1570979908108711,
      "num_tokens": 2669458.0,
      "step": 1190
    },
    {
      "entropy": 6.347487926483154,
      "epoch": 1.292049756625203,
      "grad_norm": 1.25,
      "learning_rate": 0.0004993848184509086,
      "loss": 6.1223,
      "mean_token_accuracy": 0.1580820508301258,
      "num_tokens": 2679727.0,
      "step": 1195
    },
    {
      "entropy": 6.262456750869751,
      "epoch": 1.2974580854515954,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004993527148315193,
      "loss": 6.092,
      "mean_token_accuracy": 0.15162299424409867,
      "num_tokens": 2691031.0,
      "step": 1200
    },
    {
      "entropy": 6.2036655902862545,
      "epoch": 1.3028664142779882,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004993197959138458,
      "loss": 5.8833,
      "mean_token_accuracy": 0.16101625710725784,
      "num_tokens": 2701699.0,
      "step": 1205
    },
    {
      "entropy": 6.183501052856445,
      "epoch": 1.3082747431043806,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004992860618175156,
      "loss": 5.9299,
      "mean_token_accuracy": 0.15750151723623276,
      "num_tokens": 2713038.0,
      "step": 1210
    },
    {
      "entropy": 6.1955572128295895,
      "epoch": 1.3136830719307735,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004992515126651184,
      "loss": 5.9293,
      "mean_token_accuracy": 0.1570618689060211,
      "num_tokens": 2723685.0,
      "step": 1215
    },
    {
      "entropy": 6.195014238357544,
      "epoch": 1.319091400757166,
      "grad_norm": 1.125,
      "learning_rate": 0.0004992161485822059,
      "loss": 6.0006,
      "mean_token_accuracy": 0.1587825059890747,
      "num_tokens": 2735341.0,
      "step": 1220
    },
    {
      "entropy": 6.2399842739105225,
      "epoch": 1.3244997295835588,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004991799696972913,
      "loss": 5.9313,
      "mean_token_accuracy": 0.16518269777297973,
      "num_tokens": 2746052.0,
      "step": 1225
    },
    {
      "entropy": 6.166414260864258,
      "epoch": 1.3299080584099512,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004991429761418488,
      "loss": 5.9363,
      "mean_token_accuracy": 0.15754909813404083,
      "num_tokens": 2757155.0,
      "step": 1230
    },
    {
      "entropy": 6.239412260055542,
      "epoch": 1.335316387236344,
      "grad_norm": 1.046875,
      "learning_rate": 0.000499105168050313,
      "loss": 5.9929,
      "mean_token_accuracy": 0.15189812183380128,
      "num_tokens": 2769631.0,
      "step": 1235
    },
    {
      "entropy": 6.324047040939331,
      "epoch": 1.3407247160627367,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004990665455600787,
      "loss": 6.0868,
      "mean_token_accuracy": 0.1567138358950615,
      "num_tokens": 2780797.0,
      "step": 1240
    },
    {
      "entropy": 6.166640758514404,
      "epoch": 1.3461330448891293,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004990271088114999,
      "loss": 5.946,
      "mean_token_accuracy": 0.15337988138198852,
      "num_tokens": 2792721.0,
      "step": 1245
    },
    {
      "entropy": 6.2248719215393065,
      "epoch": 1.351541373715522,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004989868579478901,
      "loss": 6.0004,
      "mean_token_accuracy": 0.16802925169467925,
      "num_tokens": 2804531.0,
      "step": 1250
    },
    {
      "entropy": 6.296425151824951,
      "epoch": 1.3569497025419146,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004989457931155209,
      "loss": 5.9965,
      "mean_token_accuracy": 0.15395874083042144,
      "num_tokens": 2816666.0,
      "step": 1255
    },
    {
      "entropy": 6.195201396942139,
      "epoch": 1.3623580313683072,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004989039144636219,
      "loss": 6.0242,
      "mean_token_accuracy": 0.15923022627830505,
      "num_tokens": 2827977.0,
      "step": 1260
    },
    {
      "entropy": 6.20063362121582,
      "epoch": 1.3677663601946999,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004988612221443804,
      "loss": 5.9936,
      "mean_token_accuracy": 0.15627091377973557,
      "num_tokens": 2839069.0,
      "step": 1265
    },
    {
      "entropy": 6.340277481079101,
      "epoch": 1.3731746890210925,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004988177163129403,
      "loss": 6.0808,
      "mean_token_accuracy": 0.15119716450572013,
      "num_tokens": 2850632.0,
      "step": 1270
    },
    {
      "entropy": 6.207772541046142,
      "epoch": 1.3785830178474852,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004987733971274018,
      "loss": 5.9607,
      "mean_token_accuracy": 0.15018116384744645,
      "num_tokens": 2862204.0,
      "step": 1275
    },
    {
      "entropy": 6.178734588623047,
      "epoch": 1.3839913466738778,
      "grad_norm": 1.125,
      "learning_rate": 0.0004987282647488211,
      "loss": 5.9197,
      "mean_token_accuracy": 0.1660077005624771,
      "num_tokens": 2874010.0,
      "step": 1280
    },
    {
      "entropy": 6.195025396347046,
      "epoch": 1.3893996755002704,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004986823193412093,
      "loss": 5.9763,
      "mean_token_accuracy": 0.16176897138357163,
      "num_tokens": 2884315.0,
      "step": 1285
    },
    {
      "entropy": 6.326874685287476,
      "epoch": 1.394808004326663,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000498635561071532,
      "loss": 6.133,
      "mean_token_accuracy": 0.15456459820270538,
      "num_tokens": 2896327.0,
      "step": 1290
    },
    {
      "entropy": 6.183460760116577,
      "epoch": 1.4002163331530557,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000498587990109709,
      "loss": 5.913,
      "mean_token_accuracy": 0.16137443631887435,
      "num_tokens": 2907007.0,
      "step": 1295
    },
    {
      "entropy": 6.183216190338134,
      "epoch": 1.4056246619794484,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004985396066286133,
      "loss": 5.9469,
      "mean_token_accuracy": 0.16008795648813248,
      "num_tokens": 2918852.0,
      "step": 1300
    },
    {
      "entropy": 6.222980594635009,
      "epoch": 1.411032990805841,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004984904108040706,
      "loss": 5.86,
      "mean_token_accuracy": 0.16177986562252045,
      "num_tokens": 2929948.0,
      "step": 1305
    },
    {
      "entropy": 6.102756023406982,
      "epoch": 1.4164413196322336,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004984404028148585,
      "loss": 5.9403,
      "mean_token_accuracy": 0.15873561054468155,
      "num_tokens": 2941299.0,
      "step": 1310
    },
    {
      "entropy": 6.112520980834961,
      "epoch": 1.4218496484586263,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004983895828427062,
      "loss": 5.8676,
      "mean_token_accuracy": 0.15320820808410646,
      "num_tokens": 2952283.0,
      "step": 1315
    },
    {
      "entropy": 6.400298118591309,
      "epoch": 1.427257977285019,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004983379510722935,
      "loss": 6.2265,
      "mean_token_accuracy": 0.15123522654175758,
      "num_tokens": 2962241.0,
      "step": 1320
    },
    {
      "entropy": 6.251223230361939,
      "epoch": 1.4326663061114115,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004982855076912506,
      "loss": 5.9829,
      "mean_token_accuracy": 0.154976050555706,
      "num_tokens": 2973408.0,
      "step": 1325
    },
    {
      "entropy": 6.191674470901489,
      "epoch": 1.4380746349378042,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004982322528901568,
      "loss": 6.0012,
      "mean_token_accuracy": 0.15962311923503875,
      "num_tokens": 2985877.0,
      "step": 1330
    },
    {
      "entropy": 6.265266704559326,
      "epoch": 1.4434829637641968,
      "grad_norm": 0.8828125,
      "learning_rate": 0.00049817818686254,
      "loss": 6.0082,
      "mean_token_accuracy": 0.15608256608247756,
      "num_tokens": 2996935.0,
      "step": 1335
    },
    {
      "entropy": 6.144498634338379,
      "epoch": 1.4488912925905895,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004981233098048765,
      "loss": 5.8997,
      "mean_token_accuracy": 0.16039617508649825,
      "num_tokens": 3007932.0,
      "step": 1340
    },
    {
      "entropy": 6.250757074356079,
      "epoch": 1.454299621416982,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004980676219165894,
      "loss": 5.9958,
      "mean_token_accuracy": 0.156630976498127,
      "num_tokens": 3019368.0,
      "step": 1345
    },
    {
      "entropy": 6.234979820251465,
      "epoch": 1.4597079502433747,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000498011123400049,
      "loss": 6.0473,
      "mean_token_accuracy": 0.15403905063867568,
      "num_tokens": 3029787.0,
      "step": 1350
    },
    {
      "entropy": 6.161543560028076,
      "epoch": 1.4651162790697674,
      "grad_norm": 1.0,
      "learning_rate": 0.0004979538144605707,
      "loss": 5.994,
      "mean_token_accuracy": 0.15943700075149536,
      "num_tokens": 3042100.0,
      "step": 1355
    },
    {
      "entropy": 6.1883018016815186,
      "epoch": 1.47052460789616,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004978956953064154,
      "loss": 6.0228,
      "mean_token_accuracy": 0.15444456189870834,
      "num_tokens": 3054160.0,
      "step": 1360
    },
    {
      "entropy": 6.235822486877441,
      "epoch": 1.4759329367225527,
      "grad_norm": 1.125,
      "learning_rate": 0.0004978367661487883,
      "loss": 5.9728,
      "mean_token_accuracy": 0.15552034378051757,
      "num_tokens": 3065401.0,
      "step": 1365
    },
    {
      "entropy": 6.124010753631592,
      "epoch": 1.4813412655489453,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004977770272018379,
      "loss": 5.915,
      "mean_token_accuracy": 0.1574919804930687,
      "num_tokens": 3076137.0,
      "step": 1370
    },
    {
      "entropy": 6.223858880996704,
      "epoch": 1.486749594375338,
      "grad_norm": 1.015625,
      "learning_rate": 0.000497716478682656,
      "loss": 6.0458,
      "mean_token_accuracy": 0.15904476791620253,
      "num_tokens": 3087274.0,
      "step": 1375
    },
    {
      "entropy": 6.195795488357544,
      "epoch": 1.4921579232017308,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004976551208112759,
      "loss": 5.9187,
      "mean_token_accuracy": 0.16271037310361863,
      "num_tokens": 3098476.0,
      "step": 1380
    },
    {
      "entropy": 6.184743976593017,
      "epoch": 1.4975662520281232,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004975929538106721,
      "loss": 5.9511,
      "mean_token_accuracy": 0.15797366350889205,
      "num_tokens": 3110060.0,
      "step": 1385
    },
    {
      "entropy": 6.053498268127441,
      "epoch": 1.502974580854516,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004975299779067598,
      "loss": 5.8576,
      "mean_token_accuracy": 0.1606665775179863,
      "num_tokens": 3121032.0,
      "step": 1390
    },
    {
      "entropy": 6.189790344238281,
      "epoch": 1.5083829096809085,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004974661933283936,
      "loss": 5.8774,
      "mean_token_accuracy": 0.16694698929786683,
      "num_tokens": 3131999.0,
      "step": 1395
    },
    {
      "entropy": 6.027636241912842,
      "epoch": 1.5137912385073014,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004974016003073667,
      "loss": 5.8154,
      "mean_token_accuracy": 0.1604975640773773,
      "num_tokens": 3142693.0,
      "step": 1400
    },
    {
      "entropy": 6.085386133193969,
      "epoch": 1.5191995673336938,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004973361990784104,
      "loss": 5.767,
      "mean_token_accuracy": 0.16097440123558043,
      "num_tokens": 3154179.0,
      "step": 1405
    },
    {
      "entropy": 6.1582213878631595,
      "epoch": 1.5246078961600866,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004972699898791927,
      "loss": 5.857,
      "mean_token_accuracy": 0.16564760953187943,
      "num_tokens": 3165314.0,
      "step": 1410
    },
    {
      "entropy": 6.119766902923584,
      "epoch": 1.530016224986479,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004972029729503182,
      "loss": 5.9687,
      "mean_token_accuracy": 0.1669076904654503,
      "num_tokens": 3177646.0,
      "step": 1415
    },
    {
      "entropy": 6.218665695190429,
      "epoch": 1.535424553812872,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004971351485353265,
      "loss": 5.9586,
      "mean_token_accuracy": 0.1630210280418396,
      "num_tokens": 3188401.0,
      "step": 1420
    },
    {
      "entropy": 6.0617822170257565,
      "epoch": 1.5408328826392643,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004970665168806917,
      "loss": 5.8469,
      "mean_token_accuracy": 0.15670444518327714,
      "num_tokens": 3200206.0,
      "step": 1425
    },
    {
      "entropy": 6.157464456558228,
      "epoch": 1.5462412114656572,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004969970782358214,
      "loss": 5.9625,
      "mean_token_accuracy": 0.1585618481040001,
      "num_tokens": 3211981.0,
      "step": 1430
    },
    {
      "entropy": 6.088303804397583,
      "epoch": 1.5516495402920496,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004969268328530557,
      "loss": 5.976,
      "mean_token_accuracy": 0.1560826152563095,
      "num_tokens": 3221997.0,
      "step": 1435
    },
    {
      "entropy": 6.2379875659942625,
      "epoch": 1.5570578691184425,
      "grad_norm": 1.0,
      "learning_rate": 0.0004968557809876666,
      "loss": 5.9227,
      "mean_token_accuracy": 0.15799657702445985,
      "num_tokens": 3234333.0,
      "step": 1440
    },
    {
      "entropy": 6.089317321777344,
      "epoch": 1.562466197944835,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004967839228978568,
      "loss": 5.9558,
      "mean_token_accuracy": 0.15238141193985938,
      "num_tokens": 3245083.0,
      "step": 1445
    },
    {
      "entropy": 6.162514019012451,
      "epoch": 1.5678745267712277,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004967112588447585,
      "loss": 5.7954,
      "mean_token_accuracy": 0.16877318769693375,
      "num_tokens": 3255787.0,
      "step": 1450
    },
    {
      "entropy": 6.064874601364136,
      "epoch": 1.5732828555976204,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004966377890924332,
      "loss": 5.9603,
      "mean_token_accuracy": 0.16299306154251098,
      "num_tokens": 3267793.0,
      "step": 1455
    },
    {
      "entropy": 6.110070848464966,
      "epoch": 1.578691184424013,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004965635139078704,
      "loss": 5.839,
      "mean_token_accuracy": 0.16499156802892684,
      "num_tokens": 3279100.0,
      "step": 1460
    },
    {
      "entropy": 6.144229125976563,
      "epoch": 1.5840995132504057,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004964884335609857,
      "loss": 6.0038,
      "mean_token_accuracy": 0.16165205985307693,
      "num_tokens": 3290251.0,
      "step": 1465
    },
    {
      "entropy": 6.1137713432312015,
      "epoch": 1.5895078420767983,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004964125483246218,
      "loss": 5.8644,
      "mean_token_accuracy": 0.16258378326892853,
      "num_tokens": 3301817.0,
      "step": 1470
    },
    {
      "entropy": 6.197260141372681,
      "epoch": 1.594916170903191,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004963358584745457,
      "loss": 5.9241,
      "mean_token_accuracy": 0.1570972815155983,
      "num_tokens": 3313435.0,
      "step": 1475
    },
    {
      "entropy": 6.134345865249633,
      "epoch": 1.6003244997295836,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004962583642894482,
      "loss": 6.0332,
      "mean_token_accuracy": 0.15797800868749617,
      "num_tokens": 3325489.0,
      "step": 1480
    },
    {
      "entropy": 6.202575826644898,
      "epoch": 1.6057328285559762,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004961800660509437,
      "loss": 5.9061,
      "mean_token_accuracy": 0.16250776648521423,
      "num_tokens": 3337030.0,
      "step": 1485
    },
    {
      "entropy": 6.009425020217895,
      "epoch": 1.6111411573823688,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000496100964043568,
      "loss": 5.8656,
      "mean_token_accuracy": 0.16609732657670975,
      "num_tokens": 3347766.0,
      "step": 1490
    },
    {
      "entropy": 6.021034860610962,
      "epoch": 1.6165494862087615,
      "grad_norm": 1.03125,
      "learning_rate": 0.000496021058554778,
      "loss": 5.7761,
      "mean_token_accuracy": 0.16601542234420777,
      "num_tokens": 3360379.0,
      "step": 1495
    },
    {
      "entropy": 6.08537859916687,
      "epoch": 1.6219578150351541,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004959403498749505,
      "loss": 5.9257,
      "mean_token_accuracy": 0.16757240146398544,
      "num_tokens": 3370962.0,
      "step": 1500
    },
    {
      "epoch": 1.6219578150351541,
      "eval_entropy": 6.020744280381636,
      "eval_loss": 6.140914440155029,
      "eval_mean_token_accuracy": 0.15917101580310952,
      "eval_num_tokens": 3370962.0,
      "eval_runtime": 1.9427,
      "eval_samples_per_second": 1809.294,
      "eval_steps_per_second": 226.483,
      "step": 1500
    },
    {
      "entropy": 6.164214992523194,
      "epoch": 1.6273661438615468,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004958588382973809,
      "loss": 5.8379,
      "mean_token_accuracy": 0.16315794885158538,
      "num_tokens": 3382648.0,
      "step": 1505
    },
    {
      "entropy": 6.041830253601074,
      "epoch": 1.6327744726879394,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004957765241182826,
      "loss": 5.9036,
      "mean_token_accuracy": 0.16003947407007219,
      "num_tokens": 3393871.0,
      "step": 1510
    },
    {
      "entropy": 6.033273839950562,
      "epoch": 1.638182801514332,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004956934076367854,
      "loss": 5.7917,
      "mean_token_accuracy": 0.16898759454488754,
      "num_tokens": 3405148.0,
      "step": 1515
    },
    {
      "entropy": 6.072603559494018,
      "epoch": 1.6435911303407247,
      "grad_norm": 1.0625,
      "learning_rate": 0.000495609489154935,
      "loss": 5.8284,
      "mean_token_accuracy": 0.17306120544672013,
      "num_tokens": 3416647.0,
      "step": 1520
    },
    {
      "entropy": 6.115102910995484,
      "epoch": 1.6489994591671173,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004955247689776911,
      "loss": 5.9205,
      "mean_token_accuracy": 0.15651903450489044,
      "num_tokens": 3428331.0,
      "step": 1525
    },
    {
      "entropy": 5.937150382995606,
      "epoch": 1.65440778799351,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000495439247412927,
      "loss": 5.776,
      "mean_token_accuracy": 0.16460416913032533,
      "num_tokens": 3440683.0,
      "step": 1530
    },
    {
      "entropy": 6.128424644470215,
      "epoch": 1.6598161168199026,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004953529247714285,
      "loss": 5.7777,
      "mean_token_accuracy": 0.16499500274658202,
      "num_tokens": 3452375.0,
      "step": 1535
    },
    {
      "entropy": 6.042714262008667,
      "epoch": 1.6652244456462952,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004952658013668922,
      "loss": 5.9321,
      "mean_token_accuracy": 0.15212839990854263,
      "num_tokens": 3463255.0,
      "step": 1540
    },
    {
      "entropy": 6.175636625289917,
      "epoch": 1.670632774472688,
      "grad_norm": 1.0,
      "learning_rate": 0.0004951778775159246,
      "loss": 5.8622,
      "mean_token_accuracy": 0.15969557911157609,
      "num_tokens": 3473642.0,
      "step": 1545
    },
    {
      "entropy": 6.021358013153076,
      "epoch": 1.6760411032990805,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004950891535380414,
      "loss": 5.8372,
      "mean_token_accuracy": 0.16012888848781587,
      "num_tokens": 3484877.0,
      "step": 1550
    },
    {
      "entropy": 6.014034366607666,
      "epoch": 1.6814494321254734,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004949996297556656,
      "loss": 5.7532,
      "mean_token_accuracy": 0.1709960624575615,
      "num_tokens": 3495500.0,
      "step": 1555
    },
    {
      "entropy": 6.051374912261963,
      "epoch": 1.6868577609518658,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004949093064941268,
      "loss": 5.8748,
      "mean_token_accuracy": 0.16371863037347795,
      "num_tokens": 3508057.0,
      "step": 1560
    },
    {
      "entropy": 6.128537130355835,
      "epoch": 1.6922660897782587,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004948181840816601,
      "loss": 6.0298,
      "mean_token_accuracy": 0.15803078263998033,
      "num_tokens": 3521110.0,
      "step": 1565
    },
    {
      "entropy": 6.136459016799927,
      "epoch": 1.697674418604651,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004947262628494045,
      "loss": 5.9607,
      "mean_token_accuracy": 0.16685472428798676,
      "num_tokens": 3530692.0,
      "step": 1570
    },
    {
      "entropy": 6.1080584049224855,
      "epoch": 1.703082747431044,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004946335431314018,
      "loss": 5.9074,
      "mean_token_accuracy": 0.15569821149110794,
      "num_tokens": 3542195.0,
      "step": 1575
    },
    {
      "entropy": 6.141083526611328,
      "epoch": 1.7084910762574363,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004945400252645959,
      "loss": 5.7995,
      "mean_token_accuracy": 0.17060298025608062,
      "num_tokens": 3552676.0,
      "step": 1580
    },
    {
      "entropy": 5.938297843933105,
      "epoch": 1.7138994050838292,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004944457095888308,
      "loss": 5.7628,
      "mean_token_accuracy": 0.16402305364608766,
      "num_tokens": 3563308.0,
      "step": 1585
    },
    {
      "entropy": 6.0929491996765135,
      "epoch": 1.7193077339102216,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00049435059644685,
      "loss": 5.7927,
      "mean_token_accuracy": 0.16668124496936798,
      "num_tokens": 3574085.0,
      "step": 1590
    },
    {
      "entropy": 6.06469030380249,
      "epoch": 1.7247160627366145,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004942546861842947,
      "loss": 5.8596,
      "mean_token_accuracy": 0.15768925547599794,
      "num_tokens": 3585816.0,
      "step": 1595
    },
    {
      "entropy": 6.0371232509613035,
      "epoch": 1.730124391563007,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004941579791497032,
      "loss": 5.9079,
      "mean_token_accuracy": 0.16382428705692292,
      "num_tokens": 3597776.0,
      "step": 1600
    },
    {
      "entropy": 6.168297672271729,
      "epoch": 1.7355327203893998,
      "grad_norm": 1.125,
      "learning_rate": 0.0004940604756945091,
      "loss": 5.9224,
      "mean_token_accuracy": 0.1619665116071701,
      "num_tokens": 3607991.0,
      "step": 1605
    },
    {
      "entropy": 6.001934623718261,
      "epoch": 1.7409410492157922,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004939621761730399,
      "loss": 5.8072,
      "mean_token_accuracy": 0.16959517300128937,
      "num_tokens": 3618204.0,
      "step": 1610
    },
    {
      "entropy": 6.072473382949829,
      "epoch": 1.746349378042185,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004938630809425166,
      "loss": 5.9081,
      "mean_token_accuracy": 0.15966459661722182,
      "num_tokens": 3629558.0,
      "step": 1615
    },
    {
      "entropy": 6.1015989780426025,
      "epoch": 1.7517577068685775,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004937631903630513,
      "loss": 5.8555,
      "mean_token_accuracy": 0.1621633917093277,
      "num_tokens": 3640773.0,
      "step": 1620
    },
    {
      "entropy": 6.01327805519104,
      "epoch": 1.7571660356949703,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004936625047976468,
      "loss": 5.7683,
      "mean_token_accuracy": 0.16852811276912688,
      "num_tokens": 3652186.0,
      "step": 1625
    },
    {
      "entropy": 5.997731494903564,
      "epoch": 1.7625743645213627,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004935610246121944,
      "loss": 5.8237,
      "mean_token_accuracy": 0.1706862360239029,
      "num_tokens": 3663068.0,
      "step": 1630
    },
    {
      "entropy": 6.129118490219116,
      "epoch": 1.7679826933477556,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004934587501754736,
      "loss": 5.9509,
      "mean_token_accuracy": 0.1577262118458748,
      "num_tokens": 3673803.0,
      "step": 1635
    },
    {
      "entropy": 6.149345588684082,
      "epoch": 1.7733910221741482,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004933556818591498,
      "loss": 5.9262,
      "mean_token_accuracy": 0.16089432835578918,
      "num_tokens": 3685482.0,
      "step": 1640
    },
    {
      "entropy": 6.019583892822266,
      "epoch": 1.7787993510005409,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004932518200377736,
      "loss": 5.7909,
      "mean_token_accuracy": 0.17309808731079102,
      "num_tokens": 3697641.0,
      "step": 1645
    },
    {
      "entropy": 6.032641172409058,
      "epoch": 1.7842076798269335,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004931471650887789,
      "loss": 5.8044,
      "mean_token_accuracy": 0.16572408378124237,
      "num_tokens": 3709560.0,
      "step": 1650
    },
    {
      "entropy": 6.101502990722656,
      "epoch": 1.7896160086533262,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004930417173924822,
      "loss": 5.9246,
      "mean_token_accuracy": 0.15817164331674577,
      "num_tokens": 3720422.0,
      "step": 1655
    },
    {
      "entropy": 5.942840242385865,
      "epoch": 1.7950243374797188,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004929354773320806,
      "loss": 5.7543,
      "mean_token_accuracy": 0.16456945538520812,
      "num_tokens": 3731104.0,
      "step": 1660
    },
    {
      "entropy": 6.045040225982666,
      "epoch": 1.8004326663061114,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004928284452936506,
      "loss": 5.9151,
      "mean_token_accuracy": 0.1589358001947403,
      "num_tokens": 3742766.0,
      "step": 1665
    },
    {
      "entropy": 6.057123613357544,
      "epoch": 1.805840995132504,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004927206216661473,
      "loss": 5.7922,
      "mean_token_accuracy": 0.16667329967021943,
      "num_tokens": 3753064.0,
      "step": 1670
    },
    {
      "entropy": 6.060616970062256,
      "epoch": 1.8112493239588967,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004926120068414015,
      "loss": 5.8339,
      "mean_token_accuracy": 0.16451720893383026,
      "num_tokens": 3765183.0,
      "step": 1675
    },
    {
      "entropy": 6.038819885253906,
      "epoch": 1.8166576527852893,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004925026012141201,
      "loss": 5.823,
      "mean_token_accuracy": 0.1619720585644245,
      "num_tokens": 3777902.0,
      "step": 1680
    },
    {
      "entropy": 5.983007335662842,
      "epoch": 1.822065981611682,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004923924051818832,
      "loss": 5.7891,
      "mean_token_accuracy": 0.1674435928463936,
      "num_tokens": 3788461.0,
      "step": 1685
    },
    {
      "entropy": 5.970801639556885,
      "epoch": 1.8274743104380746,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004922814191451435,
      "loss": 5.7166,
      "mean_token_accuracy": 0.1697210729122162,
      "num_tokens": 3799574.0,
      "step": 1690
    },
    {
      "entropy": 6.0467628002166744,
      "epoch": 1.8328826392644673,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004921696435072245,
      "loss": 5.8843,
      "mean_token_accuracy": 0.16265512108802796,
      "num_tokens": 3811889.0,
      "step": 1695
    },
    {
      "entropy": 5.9999645233154295,
      "epoch": 1.83829096809086,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004920570786743191,
      "loss": 5.8481,
      "mean_token_accuracy": 0.16382198184728622,
      "num_tokens": 3822949.0,
      "step": 1700
    },
    {
      "entropy": 6.102803945541382,
      "epoch": 1.8436992969172525,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004919437250554882,
      "loss": 5.9143,
      "mean_token_accuracy": 0.16273610591888427,
      "num_tokens": 3834734.0,
      "step": 1705
    },
    {
      "entropy": 6.029530954360962,
      "epoch": 1.8491076257436452,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004918295830626592,
      "loss": 5.7834,
      "mean_token_accuracy": 0.17444511353969575,
      "num_tokens": 3845508.0,
      "step": 1710
    },
    {
      "entropy": 5.965552425384521,
      "epoch": 1.8545159545700378,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004917146531106242,
      "loss": 5.7701,
      "mean_token_accuracy": 0.16514772921800613,
      "num_tokens": 3856500.0,
      "step": 1715
    },
    {
      "entropy": 6.14597716331482,
      "epoch": 1.8599242833964305,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004915989356170389,
      "loss": 5.8608,
      "mean_token_accuracy": 0.16108423322439194,
      "num_tokens": 3867877.0,
      "step": 1720
    },
    {
      "entropy": 6.003490257263183,
      "epoch": 1.865332612222823,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004914824310024213,
      "loss": 5.8502,
      "mean_token_accuracy": 0.1659428983926773,
      "num_tokens": 3878613.0,
      "step": 1725
    },
    {
      "entropy": 5.988573265075684,
      "epoch": 1.870740941049216,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004913651396901489,
      "loss": 5.7536,
      "mean_token_accuracy": 0.16964653134346008,
      "num_tokens": 3888418.0,
      "step": 1730
    },
    {
      "entropy": 6.0804027080535885,
      "epoch": 1.8761492698756084,
      "grad_norm": 1.046875,
      "learning_rate": 0.000491247062106459,
      "loss": 5.8341,
      "mean_token_accuracy": 0.16630043238401412,
      "num_tokens": 3899359.0,
      "step": 1735
    },
    {
      "entropy": 6.027860641479492,
      "epoch": 1.8815575987020012,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004911281986804455,
      "loss": 5.9112,
      "mean_token_accuracy": 0.16525566279888154,
      "num_tokens": 3911568.0,
      "step": 1740
    },
    {
      "entropy": 6.022890949249268,
      "epoch": 1.8869659275283936,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004910085498440587,
      "loss": 5.8767,
      "mean_token_accuracy": 0.15697558373212814,
      "num_tokens": 3921597.0,
      "step": 1745
    },
    {
      "entropy": 6.013006639480591,
      "epoch": 1.8923742563547865,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004908881160321025,
      "loss": 5.7616,
      "mean_token_accuracy": 0.1611903950572014,
      "num_tokens": 3933124.0,
      "step": 1750
    },
    {
      "entropy": 6.026111459732055,
      "epoch": 1.897782585181179,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004907668976822338,
      "loss": 5.7779,
      "mean_token_accuracy": 0.17135102599859237,
      "num_tokens": 3944388.0,
      "step": 1755
    },
    {
      "entropy": 6.03193097114563,
      "epoch": 1.9031909140075718,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004906448952349603,
      "loss": 5.7552,
      "mean_token_accuracy": 0.16768674552440643,
      "num_tokens": 3955024.0,
      "step": 1760
    },
    {
      "entropy": 5.874382495880127,
      "epoch": 1.9085992428339642,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004905221091336392,
      "loss": 5.752,
      "mean_token_accuracy": 0.17219420224428178,
      "num_tokens": 3966029.0,
      "step": 1765
    },
    {
      "entropy": 5.936220693588257,
      "epoch": 1.914007571660357,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004903985398244753,
      "loss": 5.72,
      "mean_token_accuracy": 0.16943709850311278,
      "num_tokens": 3976066.0,
      "step": 1770
    },
    {
      "entropy": 6.098024606704712,
      "epoch": 1.9194159004867495,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004902741877565201,
      "loss": 5.9178,
      "mean_token_accuracy": 0.1621759831905365,
      "num_tokens": 3986467.0,
      "step": 1775
    },
    {
      "entropy": 6.042089986801147,
      "epoch": 1.9248242293131423,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000490149053381669,
      "loss": 5.8772,
      "mean_token_accuracy": 0.16438196301460267,
      "num_tokens": 3998208.0,
      "step": 1780
    },
    {
      "entropy": 6.0255107402801515,
      "epoch": 1.9302325581395348,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004900231371546608,
      "loss": 5.8455,
      "mean_token_accuracy": 0.17020723074674607,
      "num_tokens": 4008202.0,
      "step": 1785
    },
    {
      "entropy": 5.9757099628448485,
      "epoch": 1.9356408869659276,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004898964395330753,
      "loss": 5.6556,
      "mean_token_accuracy": 0.1747577890753746,
      "num_tokens": 4019256.0,
      "step": 1790
    },
    {
      "entropy": 5.862748003005981,
      "epoch": 1.94104921579232,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004897689609773318,
      "loss": 5.7121,
      "mean_token_accuracy": 0.16850950121879577,
      "num_tokens": 4029738.0,
      "step": 1795
    },
    {
      "entropy": 6.027835559844971,
      "epoch": 1.946457544618713,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004896407019506879,
      "loss": 5.9316,
      "mean_token_accuracy": 0.15625764727592467,
      "num_tokens": 4042277.0,
      "step": 1800
    },
    {
      "entropy": 6.052269363403321,
      "epoch": 1.9518658734451053,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004895116629192371,
      "loss": 5.796,
      "mean_token_accuracy": 0.17734160125255585,
      "num_tokens": 4054019.0,
      "step": 1805
    },
    {
      "entropy": 5.959774589538574,
      "epoch": 1.9572742022714982,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004893818443519074,
      "loss": 5.8651,
      "mean_token_accuracy": 0.16139881312847137,
      "num_tokens": 4065562.0,
      "step": 1810
    },
    {
      "entropy": 6.076329278945923,
      "epoch": 1.9626825310978906,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00048925124672046,
      "loss": 5.899,
      "mean_token_accuracy": 0.16672396659851074,
      "num_tokens": 4076617.0,
      "step": 1815
    },
    {
      "entropy": 5.9678566455841064,
      "epoch": 1.9680908599242835,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004891198704994868,
      "loss": 5.7197,
      "mean_token_accuracy": 0.17168603837490082,
      "num_tokens": 4087191.0,
      "step": 1820
    },
    {
      "entropy": 6.026430892944336,
      "epoch": 1.973499188750676,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004889877161664096,
      "loss": 5.7438,
      "mean_token_accuracy": 0.17233943641185762,
      "num_tokens": 4099150.0,
      "step": 1825
    },
    {
      "entropy": 5.938602876663208,
      "epoch": 1.9789075175770687,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004888547842014771,
      "loss": 5.7592,
      "mean_token_accuracy": 0.1716069221496582,
      "num_tokens": 4110299.0,
      "step": 1830
    },
    {
      "entropy": 5.977336502075195,
      "epoch": 1.9843158464034614,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004887210750877648,
      "loss": 5.7691,
      "mean_token_accuracy": 0.1696719065308571,
      "num_tokens": 4121682.0,
      "step": 1835
    },
    {
      "entropy": 5.967773580551148,
      "epoch": 1.989724175229854,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004885865893111716,
      "loss": 5.7105,
      "mean_token_accuracy": 0.16929261684417723,
      "num_tokens": 4132914.0,
      "step": 1840
    },
    {
      "entropy": 5.943519020080567,
      "epoch": 1.9951325040562466,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004884513273604195,
      "loss": 5.817,
      "mean_token_accuracy": 0.16275771856307983,
      "num_tokens": 4144213.0,
      "step": 1845
    },
    {
      "entropy": 6.060820526546902,
      "epoch": 2.0,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004883152897270502,
      "loss": 5.8566,
      "mean_token_accuracy": 0.1669033153189553,
      "num_tokens": 4153778.0,
      "step": 1850
    },
    {
      "entropy": 5.90231032371521,
      "epoch": 2.005408328826393,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004881784769054253,
      "loss": 5.4767,
      "mean_token_accuracy": 0.17754273414611815,
      "num_tokens": 4165536.0,
      "step": 1855
    },
    {
      "entropy": 6.054047441482544,
      "epoch": 2.0108166576527853,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004880408893927226,
      "loss": 5.5205,
      "mean_token_accuracy": 0.1762312263250351,
      "num_tokens": 4176182.0,
      "step": 1860
    },
    {
      "entropy": 5.773842191696167,
      "epoch": 2.016224986479178,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004879025276889355,
      "loss": 5.5271,
      "mean_token_accuracy": 0.18644132167100907,
      "num_tokens": 4187592.0,
      "step": 1865
    },
    {
      "entropy": 5.9513507843017575,
      "epoch": 2.0216333153055706,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048776339229687077,
      "loss": 5.3477,
      "mean_token_accuracy": 0.19126913845539092,
      "num_tokens": 4198199.0,
      "step": 1870
    },
    {
      "entropy": 5.860542154312133,
      "epoch": 2.0270416441319634,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048762348372214683,
      "loss": 5.5766,
      "mean_token_accuracy": 0.17574259787797927,
      "num_tokens": 4208116.0,
      "step": 1875
    },
    {
      "entropy": 5.854842281341552,
      "epoch": 2.032449972958356,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00048748280247319165,
      "loss": 5.3452,
      "mean_token_accuracy": 0.184936061501503,
      "num_tokens": 4218569.0,
      "step": 1880
    },
    {
      "entropy": 5.84954047203064,
      "epoch": 2.0378583017847487,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004873413490612412,
      "loss": 5.4911,
      "mean_token_accuracy": 0.18496645092964173,
      "num_tokens": 4228785.0,
      "step": 1885
    },
    {
      "entropy": 5.945874834060669,
      "epoch": 2.043266630611141,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048719912400033757,
      "loss": 5.4494,
      "mean_token_accuracy": 0.18092841953039168,
      "num_tokens": 4238814.0,
      "step": 1890
    },
    {
      "entropy": 5.815927982330322,
      "epoch": 2.048674959437534,
      "grad_norm": 0.984375,
      "learning_rate": 0.000487056127807327,
      "loss": 5.4196,
      "mean_token_accuracy": 0.18144138157367706,
      "num_tokens": 4250544.0,
      "step": 1895
    },
    {
      "entropy": 5.868709659576416,
      "epoch": 2.0540832882639264,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00048691236100185795,
      "loss": 5.5407,
      "mean_token_accuracy": 0.17920881062746047,
      "num_tokens": 4261806.0,
      "step": 1900
    },
    {
      "entropy": 5.967426872253418,
      "epoch": 2.0594916170903192,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004867678241063792,
      "loss": 5.5081,
      "mean_token_accuracy": 0.1819603979587555,
      "num_tokens": 4272472.0,
      "step": 1905
    },
    {
      "entropy": 5.858438920974732,
      "epoch": 2.0648999459167117,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048662251764613833,
      "loss": 5.4782,
      "mean_token_accuracy": 0.18146095871925355,
      "num_tokens": 4283758.0,
      "step": 1910
    },
    {
      "entropy": 5.848507308959961,
      "epoch": 2.0703082747431045,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004864764421491793,
      "loss": 5.574,
      "mean_token_accuracy": 0.181258824467659,
      "num_tokens": 4295930.0,
      "step": 1915
    },
    {
      "entropy": 6.06169810295105,
      "epoch": 2.075716603569497,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004863295981463408,
      "loss": 5.498,
      "mean_token_accuracy": 0.1879578188061714,
      "num_tokens": 4306711.0,
      "step": 1920
    },
    {
      "entropy": 5.809031867980957,
      "epoch": 2.08112493239589,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004861819861712543,
      "loss": 5.4812,
      "mean_token_accuracy": 0.18434513956308365,
      "num_tokens": 4317694.0,
      "step": 1925
    },
    {
      "entropy": 5.958704710006714,
      "epoch": 2.086533261222282,
      "grad_norm": 1.125,
      "learning_rate": 0.00048603360676034203,
      "loss": 5.5586,
      "mean_token_accuracy": 0.1770416095852852,
      "num_tokens": 4329552.0,
      "step": 1930
    },
    {
      "entropy": 5.864098119735718,
      "epoch": 2.091941590048675,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048588446045281505,
      "loss": 5.4969,
      "mean_token_accuracy": 0.18100284337997435,
      "num_tokens": 4340479.0,
      "step": 1935
    },
    {
      "entropy": 5.802739858627319,
      "epoch": 2.0973499188750675,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004857345477906714,
      "loss": 5.3742,
      "mean_token_accuracy": 0.18729505985975264,
      "num_tokens": 4351147.0,
      "step": 1940
    },
    {
      "entropy": 5.726704216003418,
      "epoch": 2.1027582477014604,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048558386931869417,
      "loss": 5.3876,
      "mean_token_accuracy": 0.18800846487283707,
      "num_tokens": 4362962.0,
      "step": 1945
    },
    {
      "entropy": 5.889808416366577,
      "epoch": 2.1081665765278528,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004854324255844492,
      "loss": 5.4866,
      "mean_token_accuracy": 0.18484701812267304,
      "num_tokens": 4373723.0,
      "step": 1950
    },
    {
      "entropy": 5.75990924835205,
      "epoch": 2.1135749053542456,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00048528021713828336,
      "loss": 5.4655,
      "mean_token_accuracy": 0.1776862010359764,
      "num_tokens": 4383434.0,
      "step": 1955
    },
    {
      "entropy": 5.841428995132446,
      "epoch": 2.118983234180638,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004851272445333225,
      "loss": 5.4186,
      "mean_token_accuracy": 0.18994414508342744,
      "num_tokens": 4394847.0,
      "step": 1960
    },
    {
      "entropy": 5.8689299583435055,
      "epoch": 2.124391563007031,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048497350832546937,
      "loss": 5.524,
      "mean_token_accuracy": 0.18320200741291046,
      "num_tokens": 4406367.0,
      "step": 1965
    },
    {
      "entropy": 5.917090654373169,
      "epoch": 2.1297998918334233,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004848190090734018,
      "loss": 5.5273,
      "mean_token_accuracy": 0.18322407305240632,
      "num_tokens": 4417636.0,
      "step": 1970
    },
    {
      "entropy": 5.863876390457153,
      "epoch": 2.135208220659816,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048466374733857035,
      "loss": 5.5325,
      "mean_token_accuracy": 0.18607548624277115,
      "num_tokens": 4428473.0,
      "step": 1975
    },
    {
      "entropy": 5.845749950408935,
      "epoch": 2.1406165494862086,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004845077236851966,
      "loss": 5.4622,
      "mean_token_accuracy": 0.18125550746917723,
      "num_tokens": 4439780.0,
      "step": 1980
    },
    {
      "entropy": 5.8221944808959964,
      "epoch": 2.1460248783126015,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048435093868027076,
      "loss": 5.4118,
      "mean_token_accuracy": 0.19358843713998794,
      "num_tokens": 4450929.0,
      "step": 1985
    },
    {
      "entropy": 5.835688257217408,
      "epoch": 2.151433207138994,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048419339289354995,
      "loss": 5.454,
      "mean_token_accuracy": 0.18442270159721375,
      "num_tokens": 4462967.0,
      "step": 1990
    },
    {
      "entropy": 5.889276933670044,
      "epoch": 2.1568415359653867,
      "grad_norm": 1.140625,
      "learning_rate": 0.000484035086897556,
      "loss": 5.5347,
      "mean_token_accuracy": 0.1838492140173912,
      "num_tokens": 4473970.0,
      "step": 1995
    },
    {
      "entropy": 5.863001918792724,
      "epoch": 2.162249864791779,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004838760212675732,
      "loss": 5.5383,
      "mean_token_accuracy": 0.1813681200146675,
      "num_tokens": 4485660.0,
      "step": 2000
    },
    {
      "epoch": 2.162249864791779,
      "eval_entropy": 5.725599083033475,
      "eval_loss": 6.020395755767822,
      "eval_mean_token_accuracy": 0.16601560962471096,
      "eval_num_tokens": 4485660.0,
      "eval_runtime": 1.9373,
      "eval_samples_per_second": 1814.336,
      "eval_steps_per_second": 227.115,
      "step": 2000
    },
    {
      "entropy": 5.8430750370025635,
      "epoch": 2.167658193618172,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004837161965816464,
      "loss": 5.509,
      "mean_token_accuracy": 0.1867072880268097,
      "num_tokens": 4497261.0,
      "step": 2005
    },
    {
      "entropy": 5.846252346038819,
      "epoch": 2.1730665224445644,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048355561342057886,
      "loss": 5.3884,
      "mean_token_accuracy": 0.18988969624042512,
      "num_tokens": 4508120.0,
      "step": 2010
    },
    {
      "entropy": 5.839772653579712,
      "epoch": 2.1784748512709573,
      "grad_norm": 1.125,
      "learning_rate": 0.00048339427236793024,
      "loss": 5.4781,
      "mean_token_accuracy": 0.18922811150550842,
      "num_tokens": 4519286.0,
      "step": 2015
    },
    {
      "entropy": 5.898039484024048,
      "epoch": 2.1838831800973497,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004832321740100143,
      "loss": 5.5553,
      "mean_token_accuracy": 0.1808038905262947,
      "num_tokens": 4531912.0,
      "step": 2020
    },
    {
      "entropy": 5.885161018371582,
      "epoch": 2.1892915089237426,
      "grad_norm": 1.15625,
      "learning_rate": 0.00048306931893589666,
      "loss": 5.5665,
      "mean_token_accuracy": 0.18202104717493056,
      "num_tokens": 4544370.0,
      "step": 2025
    },
    {
      "entropy": 5.853564596176147,
      "epoch": 2.194699837750135,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048290570773739327,
      "loss": 5.5119,
      "mean_token_accuracy": 0.18240146934986115,
      "num_tokens": 4556013.0,
      "step": 2030
    },
    {
      "entropy": 5.764708852767944,
      "epoch": 2.200108166576528,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048274134100906747,
      "loss": 5.4625,
      "mean_token_accuracy": 0.18581133484840393,
      "num_tokens": 4566200.0,
      "step": 2035
    },
    {
      "entropy": 5.856343841552734,
      "epoch": 2.2055164954029207,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048257621934822835,
      "loss": 5.506,
      "mean_token_accuracy": 0.18467539846897124,
      "num_tokens": 4576899.0,
      "step": 2040
    },
    {
      "entropy": 5.934852552413941,
      "epoch": 2.210924824229313,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004824103433549285,
      "loss": 5.5254,
      "mean_token_accuracy": 0.17869499921798707,
      "num_tokens": 4588983.0,
      "step": 2045
    },
    {
      "entropy": 5.847223997116089,
      "epoch": 2.2163331530557056,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004822437136319616,
      "loss": 5.479,
      "mean_token_accuracy": 0.19166221618652343,
      "num_tokens": 4599758.0,
      "step": 2050
    },
    {
      "entropy": 5.81277289390564,
      "epoch": 2.2217414818820984,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004820763307848604,
      "loss": 5.5154,
      "mean_token_accuracy": 0.1851065754890442,
      "num_tokens": 4610927.0,
      "step": 2055
    },
    {
      "entropy": 5.839804220199585,
      "epoch": 2.2271498107084913,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048190819542189456,
      "loss": 5.5307,
      "mean_token_accuracy": 0.18261638432741165,
      "num_tokens": 4623719.0,
      "step": 2060
    },
    {
      "entropy": 5.836520338058472,
      "epoch": 2.2325581395348837,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00048173930815406847,
      "loss": 5.5893,
      "mean_token_accuracy": 0.17861880511045455,
      "num_tokens": 4635389.0,
      "step": 2065
    },
    {
      "entropy": 5.938038301467896,
      "epoch": 2.2379664683612766,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004815696695951187,
      "loss": 5.5353,
      "mean_token_accuracy": 0.18332355469465256,
      "num_tokens": 4645093.0,
      "step": 2070
    },
    {
      "entropy": 5.799956464767456,
      "epoch": 2.243374797187669,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004813992803615123,
      "loss": 5.5249,
      "mean_token_accuracy": 0.17770509719848632,
      "num_tokens": 4656490.0,
      "step": 2075
    },
    {
      "entropy": 5.897887754440307,
      "epoch": 2.248783126014062,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00048122814107244414,
      "loss": 5.5365,
      "mean_token_accuracy": 0.18031222522258758,
      "num_tokens": 4667603.0,
      "step": 2080
    },
    {
      "entropy": 5.810691595077515,
      "epoch": 2.2541914548404542,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048105625234983466,
      "loss": 5.4848,
      "mean_token_accuracy": 0.17508939504623414,
      "num_tokens": 4678669.0,
      "step": 2085
    },
    {
      "entropy": 5.735423612594604,
      "epoch": 2.259599783666847,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00048088361481832807,
      "loss": 5.2923,
      "mean_token_accuracy": 0.19484480917453767,
      "num_tokens": 4689757.0,
      "step": 2090
    },
    {
      "entropy": 5.711760234832764,
      "epoch": 2.2650081124932395,
      "grad_norm": 1.140625,
      "learning_rate": 0.00048071022910528934,
      "loss": 5.4666,
      "mean_token_accuracy": 0.18362647742033006,
      "num_tokens": 4699332.0,
      "step": 2095
    },
    {
      "entropy": 5.8964282989501955,
      "epoch": 2.2704164413196324,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00048053609584080274,
      "loss": 5.5409,
      "mean_token_accuracy": 0.1857800304889679,
      "num_tokens": 4709485.0,
      "step": 2100
    },
    {
      "entropy": 5.778720140457153,
      "epoch": 2.275824770146025,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048036121565766876,
      "loss": 5.4826,
      "mean_token_accuracy": 0.18560403287410737,
      "num_tokens": 4720935.0,
      "step": 2105
    },
    {
      "entropy": 5.745053100585937,
      "epoch": 2.2812330989724177,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004801855891914026,
      "loss": 5.4299,
      "mean_token_accuracy": 0.18251365274190903,
      "num_tokens": 4731219.0,
      "step": 2110
    },
    {
      "entropy": 5.898935747146607,
      "epoch": 2.28664142779881,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048000921708023117,
      "loss": 5.5586,
      "mean_token_accuracy": 0.18378493636846543,
      "num_tokens": 4742637.0,
      "step": 2115
    },
    {
      "entropy": 5.783941411972046,
      "epoch": 2.292049756625203,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004798320999650911,
      "loss": 5.4516,
      "mean_token_accuracy": 0.1839448481798172,
      "num_tokens": 4753341.0,
      "step": 2120
    },
    {
      "entropy": 5.984134101867676,
      "epoch": 2.2974580854515954,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047965423848962636,
      "loss": 5.6041,
      "mean_token_accuracy": 0.18355586379766464,
      "num_tokens": 4765339.0,
      "step": 2125
    },
    {
      "entropy": 5.843023586273193,
      "epoch": 2.302866414277988,
      "grad_norm": 1.015625,
      "learning_rate": 0.000479475633300186,
      "loss": 5.4551,
      "mean_token_accuracy": 0.18163464665412904,
      "num_tokens": 4776149.0,
      "step": 2130
    },
    {
      "entropy": 5.771022653579712,
      "epoch": 2.3082747431043806,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047929628504582173,
      "loss": 5.4777,
      "mean_token_accuracy": 0.18727335333824158,
      "num_tokens": 4788120.0,
      "step": 2135
    },
    {
      "entropy": 5.860320329666138,
      "epoch": 2.3136830719307735,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004791161943782856,
      "loss": 5.5576,
      "mean_token_accuracy": 0.17805030047893525,
      "num_tokens": 4799922.0,
      "step": 2140
    },
    {
      "entropy": 5.718722105026245,
      "epoch": 2.319091400757166,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047893536195202726,
      "loss": 5.4895,
      "mean_token_accuracy": 0.18036262691020966,
      "num_tokens": 4810501.0,
      "step": 2145
    },
    {
      "entropy": 5.817003440856934,
      "epoch": 2.3244997295835588,
      "grad_norm": 1.125,
      "learning_rate": 0.0004787537884241923,
      "loss": 5.4417,
      "mean_token_accuracy": 0.1868960365653038,
      "num_tokens": 4820919.0,
      "step": 2150
    },
    {
      "entropy": 5.889329147338867,
      "epoch": 2.329908058409951,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047857147445461943,
      "loss": 5.5355,
      "mean_token_accuracy": 0.18348777890205384,
      "num_tokens": 4833298.0,
      "step": 2155
    },
    {
      "entropy": 5.682007360458374,
      "epoch": 2.335316387236344,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004783884207058379,
      "loss": 5.4153,
      "mean_token_accuracy": 0.17922882437705995,
      "num_tokens": 4845199.0,
      "step": 2160
    },
    {
      "entropy": 5.795744609832764,
      "epoch": 2.3407247160627365,
      "grad_norm": 1.125,
      "learning_rate": 0.00047820462784306555,
      "loss": 5.4691,
      "mean_token_accuracy": 0.18276022076606752,
      "num_tokens": 4856549.0,
      "step": 2165
    },
    {
      "entropy": 5.809813499450684,
      "epoch": 2.3461330448891293,
      "grad_norm": 1.09375,
      "learning_rate": 0.000478020096534206,
      "loss": 5.5131,
      "mean_token_accuracy": 0.17976404577493668,
      "num_tokens": 4868446.0,
      "step": 2170
    },
    {
      "entropy": 5.885415935516358,
      "epoch": 2.3515413737155217,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047783482744984644,
      "loss": 5.5668,
      "mean_token_accuracy": 0.1731751322746277,
      "num_tokens": 4879047.0,
      "step": 2175
    },
    {
      "entropy": 5.818225002288818,
      "epoch": 2.3569497025419146,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00047764882126325527,
      "loss": 5.4639,
      "mean_token_accuracy": 0.18213903456926345,
      "num_tokens": 4891591.0,
      "step": 2180
    },
    {
      "entropy": 5.744760799407959,
      "epoch": 2.362358031368307,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004774620786503793,
      "loss": 5.4759,
      "mean_token_accuracy": 0.18766329139471055,
      "num_tokens": 4902710.0,
      "step": 2185
    },
    {
      "entropy": 5.887608385086059,
      "epoch": 2.3677663601947,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004772746002898417,
      "loss": 5.5475,
      "mean_token_accuracy": 0.18340907692909242,
      "num_tokens": 4914668.0,
      "step": 2190
    },
    {
      "entropy": 5.906754684448242,
      "epoch": 2.3731746890210923,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047708638686293903,
      "loss": 5.6686,
      "mean_token_accuracy": 0.17607683539390565,
      "num_tokens": 4926684.0,
      "step": 2195
    },
    {
      "entropy": 5.885916709899902,
      "epoch": 2.378583017847485,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00047689743905363963,
      "loss": 5.5433,
      "mean_token_accuracy": 0.1780522108078003,
      "num_tokens": 4938577.0,
      "step": 2200
    },
    {
      "entropy": 5.711051607131958,
      "epoch": 2.383991346673878,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004767077575485801,
      "loss": 5.4213,
      "mean_token_accuracy": 0.19123190641403198,
      "num_tokens": 4949070.0,
      "step": 2205
    },
    {
      "entropy": 5.815402317047119,
      "epoch": 2.3893996755002704,
      "grad_norm": 1.125,
      "learning_rate": 0.0004765173430370635,
      "loss": 5.4939,
      "mean_token_accuracy": 0.18558672815561295,
      "num_tokens": 4959558.0,
      "step": 2210
    },
    {
      "entropy": 5.715311431884766,
      "epoch": 2.394808004326663,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004763261962110567,
      "loss": 5.3557,
      "mean_token_accuracy": 0.19049575328826904,
      "num_tokens": 4970310.0,
      "step": 2215
    },
    {
      "entropy": 5.725232744216919,
      "epoch": 2.4002163331530557,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004761343177651875,
      "loss": 5.4447,
      "mean_token_accuracy": 0.18282105326652526,
      "num_tokens": 4982300.0,
      "step": 2220
    },
    {
      "entropy": 5.790338802337646,
      "epoch": 2.4056246619794486,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004759417083967428,
      "loss": 5.4725,
      "mean_token_accuracy": 0.18086550235748292,
      "num_tokens": 4994425.0,
      "step": 2225
    },
    {
      "entropy": 5.776576137542724,
      "epoch": 2.411032990805841,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00047574836880566544,
      "loss": 5.4794,
      "mean_token_accuracy": 0.19098425656557083,
      "num_tokens": 5005872.0,
      "step": 2230
    },
    {
      "entropy": 5.78202633857727,
      "epoch": 2.4164413196322334,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004755542996945519,
      "loss": 5.5075,
      "mean_token_accuracy": 0.18447682559490203,
      "num_tokens": 5016903.0,
      "step": 2235
    },
    {
      "entropy": 5.836867189407348,
      "epoch": 2.4218496484586263,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047535950176864976,
      "loss": 5.5442,
      "mean_token_accuracy": 0.18406234681606293,
      "num_tokens": 5028407.0,
      "step": 2240
    },
    {
      "entropy": 5.820690727233886,
      "epoch": 2.427257977285019,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047516397573585525,
      "loss": 5.5518,
      "mean_token_accuracy": 0.18487397730350494,
      "num_tokens": 5039693.0,
      "step": 2245
    },
    {
      "entropy": 5.846407175064087,
      "epoch": 2.4326663061114115,
      "grad_norm": 1.125,
      "learning_rate": 0.00047496772230671034,
      "loss": 5.5034,
      "mean_token_accuracy": 0.18430618047714234,
      "num_tokens": 5051279.0,
      "step": 2250
    },
    {
      "entropy": 5.8105754375457765,
      "epoch": 2.4380746349378044,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004747707421944004,
      "loss": 5.4522,
      "mean_token_accuracy": 0.1847480833530426,
      "num_tokens": 5062698.0,
      "step": 2255
    },
    {
      "entropy": 5.7512526512146,
      "epoch": 2.443482963764197,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004745730361147517,
      "loss": 5.4366,
      "mean_token_accuracy": 0.18818385004997254,
      "num_tokens": 5074927.0,
      "step": 2260
    },
    {
      "entropy": 5.796603631973267,
      "epoch": 2.4488912925905897,
      "grad_norm": 1.125,
      "learning_rate": 0.0004743746047862286,
      "loss": 5.5697,
      "mean_token_accuracy": 0.1745971441268921,
      "num_tokens": 5086298.0,
      "step": 2265
    },
    {
      "entropy": 5.817357540130615,
      "epoch": 2.454299621416982,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004741754489299309,
      "loss": 5.5101,
      "mean_token_accuracy": 0.17972173988819123,
      "num_tokens": 5098161.0,
      "step": 2270
    },
    {
      "entropy": 5.774695014953613,
      "epoch": 2.459707950243375,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047397556926959155,
      "loss": 5.4874,
      "mean_token_accuracy": 0.19200887233018876,
      "num_tokens": 5110743.0,
      "step": 2275
    },
    {
      "entropy": 5.90885591506958,
      "epoch": 2.4651162790697674,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047377496653157374,
      "loss": 5.5304,
      "mean_token_accuracy": 0.1880877986550331,
      "num_tokens": 5121838.0,
      "step": 2280
    },
    {
      "entropy": 5.696847534179687,
      "epoch": 2.4705246078961602,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047357364144486806,
      "loss": 5.4768,
      "mean_token_accuracy": 0.1817433074116707,
      "num_tokens": 5135047.0,
      "step": 2285
    },
    {
      "entropy": 5.772260761260986,
      "epoch": 2.4759329367225527,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047337159474109057,
      "loss": 5.4649,
      "mean_token_accuracy": 0.17703176736831666,
      "num_tokens": 5146329.0,
      "step": 2290
    },
    {
      "entropy": 5.863223314285278,
      "epoch": 2.4813412655489455,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047316882715447934,
      "loss": 5.5062,
      "mean_token_accuracy": 0.1872244283556938,
      "num_tokens": 5157511.0,
      "step": 2295
    },
    {
      "entropy": 5.693565464019775,
      "epoch": 2.486749594375338,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004729653394218922,
      "loss": 5.495,
      "mean_token_accuracy": 0.18242276459932327,
      "num_tokens": 5168456.0,
      "step": 2300
    },
    {
      "entropy": 5.931699275970459,
      "epoch": 2.492157923201731,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047276113228280427,
      "loss": 5.5515,
      "mean_token_accuracy": 0.1825096696615219,
      "num_tokens": 5179294.0,
      "step": 2305
    },
    {
      "entropy": 5.807642078399658,
      "epoch": 2.497566252028123,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004725562064793045,
      "loss": 5.4993,
      "mean_token_accuracy": 0.18253640085458755,
      "num_tokens": 5190195.0,
      "step": 2310
    },
    {
      "entropy": 5.710082483291626,
      "epoch": 2.502974580854516,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047235056275609397,
      "loss": 5.3895,
      "mean_token_accuracy": 0.18921253234148025,
      "num_tokens": 5201289.0,
      "step": 2315
    },
    {
      "entropy": 5.741735363006592,
      "epoch": 2.5083829096809085,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004721442018604822,
      "loss": 5.4177,
      "mean_token_accuracy": 0.19280524998903276,
      "num_tokens": 5212295.0,
      "step": 2320
    },
    {
      "entropy": 5.650020360946655,
      "epoch": 2.5137912385073014,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004719371245423853,
      "loss": 5.3894,
      "mean_token_accuracy": 0.1905950501561165,
      "num_tokens": 5223144.0,
      "step": 2325
    },
    {
      "entropy": 5.713900184631347,
      "epoch": 2.5191995673336938,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047172933155432273,
      "loss": 5.4097,
      "mean_token_accuracy": 0.18740693479776382,
      "num_tokens": 5232840.0,
      "step": 2330
    },
    {
      "entropy": 5.713705444335938,
      "epoch": 2.5246078961600866,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047152082365141466,
      "loss": 5.5167,
      "mean_token_accuracy": 0.18210788518190385,
      "num_tokens": 5243796.0,
      "step": 2335
    },
    {
      "entropy": 5.8688812255859375,
      "epoch": 2.530016224986479,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004713116015913794,
      "loss": 5.5962,
      "mean_token_accuracy": 0.17875071614980698,
      "num_tokens": 5255990.0,
      "step": 2340
    },
    {
      "entropy": 5.84153995513916,
      "epoch": 2.535424553812872,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00047110166613453016,
      "loss": 5.5344,
      "mean_token_accuracy": 0.1801824688911438,
      "num_tokens": 5268134.0,
      "step": 2345
    },
    {
      "entropy": 5.672592115402222,
      "epoch": 2.5408328826392643,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047089101804377305,
      "loss": 5.3767,
      "mean_token_accuracy": 0.18733149021863937,
      "num_tokens": 5278662.0,
      "step": 2350
    },
    {
      "entropy": 5.775094842910766,
      "epoch": 2.546241211465657,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004706796580846037,
      "loss": 5.4486,
      "mean_token_accuracy": 0.18052454739809037,
      "num_tokens": 5288382.0,
      "step": 2355
    },
    {
      "entropy": 5.735822057723999,
      "epoch": 2.5516495402920496,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047046758702510465,
      "loss": 5.5175,
      "mean_token_accuracy": 0.17908523827791215,
      "num_tokens": 5299769.0,
      "step": 2360
    },
    {
      "entropy": 5.894213485717773,
      "epoch": 2.5570578691184425,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047025480563594264,
      "loss": 5.4694,
      "mean_token_accuracy": 0.18954675197601317,
      "num_tokens": 5310662.0,
      "step": 2365
    },
    {
      "entropy": 5.68624210357666,
      "epoch": 2.5624661979448353,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047004131469036576,
      "loss": 5.433,
      "mean_token_accuracy": 0.1791605308651924,
      "num_tokens": 5321785.0,
      "step": 2370
    },
    {
      "entropy": 5.790686321258545,
      "epoch": 2.5678745267712277,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046982711496420057,
      "loss": 5.4881,
      "mean_token_accuracy": 0.17994098961353303,
      "num_tokens": 5333557.0,
      "step": 2375
    },
    {
      "entropy": 5.7890860080719,
      "epoch": 2.57328285559762,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046961220723584925,
      "loss": 5.5283,
      "mean_token_accuracy": 0.1812247097492218,
      "num_tokens": 5344767.0,
      "step": 2380
    },
    {
      "entropy": 5.777838706970215,
      "epoch": 2.578691184424013,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00046939659228628725,
      "loss": 5.4417,
      "mean_token_accuracy": 0.18691411167383193,
      "num_tokens": 5356673.0,
      "step": 2385
    },
    {
      "entropy": 5.7575544834136965,
      "epoch": 2.584099513250406,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046918027089905956,
      "loss": 5.5538,
      "mean_token_accuracy": 0.1779048413038254,
      "num_tokens": 5367296.0,
      "step": 2390
    },
    {
      "entropy": 5.7355200290679935,
      "epoch": 2.5895078420767983,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046896324386027873,
      "loss": 5.4201,
      "mean_token_accuracy": 0.18585974872112274,
      "num_tokens": 5378081.0,
      "step": 2395
    },
    {
      "entropy": 5.821057653427124,
      "epoch": 2.5949161709031907,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046874551195862145,
      "loss": 5.5102,
      "mean_token_accuracy": 0.17801062613725663,
      "num_tokens": 5390244.0,
      "step": 2400
    },
    {
      "entropy": 5.708388137817383,
      "epoch": 2.6003244997295836,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000468527075985326,
      "loss": 5.3888,
      "mean_token_accuracy": 0.1883876532316208,
      "num_tokens": 5400741.0,
      "step": 2405
    },
    {
      "entropy": 5.796988010406494,
      "epoch": 2.6057328285559764,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046830793673418924,
      "loss": 5.5663,
      "mean_token_accuracy": 0.18608431220054628,
      "num_tokens": 5411440.0,
      "step": 2410
    },
    {
      "entropy": 5.759448432922364,
      "epoch": 2.611141157382369,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046808809500156355,
      "loss": 5.4269,
      "mean_token_accuracy": 0.18147926032543182,
      "num_tokens": 5422472.0,
      "step": 2415
    },
    {
      "entropy": 5.705766773223877,
      "epoch": 2.6165494862087613,
      "grad_norm": 1.0,
      "learning_rate": 0.0004678675515863545,
      "loss": 5.5039,
      "mean_token_accuracy": 0.17874256670475006,
      "num_tokens": 5433958.0,
      "step": 2420
    },
    {
      "entropy": 5.767446708679199,
      "epoch": 2.621957815035154,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004676463072900173,
      "loss": 5.4192,
      "mean_token_accuracy": 0.19382819086313247,
      "num_tokens": 5444343.0,
      "step": 2425
    },
    {
      "entropy": 5.681805276870728,
      "epoch": 2.627366143861547,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00046742436291655415,
      "loss": 5.4357,
      "mean_token_accuracy": 0.18407043516635896,
      "num_tokens": 5456260.0,
      "step": 2430
    },
    {
      "entropy": 5.748001480102539,
      "epoch": 2.6327744726879394,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00046720171927251155,
      "loss": 5.5282,
      "mean_token_accuracy": 0.18753280341625214,
      "num_tokens": 5468281.0,
      "step": 2435
    },
    {
      "entropy": 5.655755233764649,
      "epoch": 2.638182801514332,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004669783771669769,
      "loss": 5.3615,
      "mean_token_accuracy": 0.19994893074035644,
      "num_tokens": 5480068.0,
      "step": 2440
    },
    {
      "entropy": 5.671969795227051,
      "epoch": 2.6435911303407247,
      "grad_norm": 1.09375,
      "learning_rate": 0.000466754337411576,
      "loss": 5.3704,
      "mean_token_accuracy": 0.18418589234352112,
      "num_tokens": 5490932.0,
      "step": 2445
    },
    {
      "entropy": 5.653346681594849,
      "epoch": 2.6489994591671175,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046652960082046985,
      "loss": 5.4672,
      "mean_token_accuracy": 0.18149202167987824,
      "num_tokens": 5501787.0,
      "step": 2450
    },
    {
      "entropy": 5.773309183120728,
      "epoch": 2.65440778799351,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046630416821035164,
      "loss": 5.4492,
      "mean_token_accuracy": 0.18520426899194717,
      "num_tokens": 5513297.0,
      "step": 2455
    },
    {
      "entropy": 5.655526876449585,
      "epoch": 2.6598161168199024,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004660780404004441,
      "loss": 5.4302,
      "mean_token_accuracy": 0.19071539640426635,
      "num_tokens": 5523797.0,
      "step": 2460
    },
    {
      "entropy": 5.685646867752075,
      "epoch": 2.6652244456462952,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046585121821249623,
      "loss": 5.4664,
      "mean_token_accuracy": 0.191667777299881,
      "num_tokens": 5535808.0,
      "step": 2465
    },
    {
      "entropy": 5.8277747631073,
      "epoch": 2.670632774472688,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00046562370247078025,
      "loss": 5.5487,
      "mean_token_accuracy": 0.1725021183490753,
      "num_tokens": 5548179.0,
      "step": 2470
    },
    {
      "entropy": 5.832285785675049,
      "epoch": 2.6760411032990805,
      "grad_norm": 1.0,
      "learning_rate": 0.000465395494002089,
      "loss": 5.5629,
      "mean_token_accuracy": 0.17875450998544692,
      "num_tokens": 5559144.0,
      "step": 2475
    },
    {
      "entropy": 5.757518625259399,
      "epoch": 2.6814494321254734,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004651665936357325,
      "loss": 5.4518,
      "mean_token_accuracy": 0.1852279335260391,
      "num_tokens": 5571103.0,
      "step": 2480
    },
    {
      "entropy": 5.690479469299317,
      "epoch": 2.686857760951866,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004649370022035354,
      "loss": 5.3842,
      "mean_token_accuracy": 0.18641475737094879,
      "num_tokens": 5583239.0,
      "step": 2485
    },
    {
      "entropy": 5.68533730506897,
      "epoch": 2.6922660897782587,
      "grad_norm": 1.125,
      "learning_rate": 0.0004647067205398332,
      "loss": 5.3563,
      "mean_token_accuracy": 0.18859488219022752,
      "num_tokens": 5593602.0,
      "step": 2490
    },
    {
      "entropy": 5.8065142154693605,
      "epoch": 2.697674418604651,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046447574948147027,
      "loss": 5.5152,
      "mean_token_accuracy": 0.17736031115055084,
      "num_tokens": 5604696.0,
      "step": 2495
    },
    {
      "entropy": 5.704569149017334,
      "epoch": 2.703082747431044,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046424408986779587,
      "loss": 5.49,
      "mean_token_accuracy": 0.17822713404893875,
      "num_tokens": 5615668.0,
      "step": 2500
    },
    {
      "epoch": 2.703082747431044,
      "eval_entropy": 5.6619423313574355,
      "eval_loss": 5.923056125640869,
      "eval_mean_token_accuracy": 0.17251565199006688,
      "eval_num_tokens": 5615668.0,
      "eval_runtime": 1.9309,
      "eval_samples_per_second": 1820.364,
      "eval_steps_per_second": 227.869,
      "step": 2500
    },
    {
      "entropy": 5.76570315361023,
      "epoch": 2.7084910762574363,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004640117425406615,
      "loss": 5.4237,
      "mean_token_accuracy": 0.18574815541505812,
      "num_tokens": 5627344.0,
      "step": 2505
    },
    {
      "entropy": 5.701715326309204,
      "epoch": 2.713899405083829,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000463778708344418,
      "loss": 5.4671,
      "mean_token_accuracy": 0.18088625818490983,
      "num_tokens": 5637472.0,
      "step": 2510
    },
    {
      "entropy": 5.713341093063354,
      "epoch": 2.7193077339102216,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046354498812591207,
      "loss": 5.4502,
      "mean_token_accuracy": 0.18748428374528886,
      "num_tokens": 5647766.0,
      "step": 2515
    },
    {
      "entropy": 5.655908727645874,
      "epoch": 2.7247160627366145,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046331058273448364,
      "loss": 5.3821,
      "mean_token_accuracy": 0.1837076410651207,
      "num_tokens": 5658343.0,
      "step": 2520
    },
    {
      "entropy": 5.677487134933472,
      "epoch": 2.730124391563007,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004630754930219624,
      "loss": 5.3865,
      "mean_token_accuracy": 0.19113706946372985,
      "num_tokens": 5669640.0,
      "step": 2525
    },
    {
      "entropy": 5.602105569839478,
      "epoch": 2.7355327203893998,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004628397198426648,
      "loss": 5.3608,
      "mean_token_accuracy": 0.18334166407585145,
      "num_tokens": 5682421.0,
      "step": 2530
    },
    {
      "entropy": 5.6781334400177,
      "epoch": 2.740941049215792,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046260326405339126,
      "loss": 5.3426,
      "mean_token_accuracy": 0.19388457536697387,
      "num_tokens": 5692797.0,
      "step": 2535
    },
    {
      "entropy": 5.761030578613282,
      "epoch": 2.746349378042185,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004623661265134226,
      "loss": 5.5239,
      "mean_token_accuracy": 0.18626205921173095,
      "num_tokens": 5702779.0,
      "step": 2540
    },
    {
      "entropy": 5.738360166549683,
      "epoch": 2.7517577068685775,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004621283080845172,
      "loss": 5.5032,
      "mean_token_accuracy": 0.18362185806035997,
      "num_tokens": 5714147.0,
      "step": 2545
    },
    {
      "entropy": 5.640808868408203,
      "epoch": 2.7571660356949703,
      "grad_norm": 0.9375,
      "learning_rate": 0.00046188980963090783,
      "loss": 5.3324,
      "mean_token_accuracy": 0.19274002760648729,
      "num_tokens": 5725718.0,
      "step": 2550
    },
    {
      "entropy": 5.647945594787598,
      "epoch": 2.7625743645213627,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046165063201929825,
      "loss": 5.4152,
      "mean_token_accuracy": 0.1916851669549942,
      "num_tokens": 5737441.0,
      "step": 2555
    },
    {
      "entropy": 5.7787117004394535,
      "epoch": 2.7679826933477556,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046141077611886074,
      "loss": 5.5182,
      "mean_token_accuracy": 0.1864737704396248,
      "num_tokens": 5748735.0,
      "step": 2560
    },
    {
      "entropy": 5.815402412414551,
      "epoch": 2.773391022174148,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004611702428012319,
      "loss": 5.5785,
      "mean_token_accuracy": 0.17667905241250992,
      "num_tokens": 5760262.0,
      "step": 2565
    },
    {
      "entropy": 5.7044572830200195,
      "epoch": 2.778799351000541,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046092903294051057,
      "loss": 5.4725,
      "mean_token_accuracy": 0.18194340616464616,
      "num_tokens": 5771909.0,
      "step": 2570
    },
    {
      "entropy": 5.587724304199218,
      "epoch": 2.7842076798269337,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046068714741325385,
      "loss": 5.371,
      "mean_token_accuracy": 0.1913776785135269,
      "num_tokens": 5782028.0,
      "step": 2575
    },
    {
      "entropy": 5.821951818466187,
      "epoch": 2.789616008653326,
      "grad_norm": 1.125,
      "learning_rate": 0.0004604445870984743,
      "loss": 5.5162,
      "mean_token_accuracy": 0.18275947123765945,
      "num_tokens": 5793641.0,
      "step": 2580
    },
    {
      "entropy": 5.798354768753052,
      "epoch": 2.7950243374797186,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004602013528776367,
      "loss": 5.4623,
      "mean_token_accuracy": 0.18469780087471008,
      "num_tokens": 5805392.0,
      "step": 2585
    },
    {
      "entropy": 5.736293888092041,
      "epoch": 2.8004326663061114,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004599574456346548,
      "loss": 5.4405,
      "mean_token_accuracy": 0.1861585184931755,
      "num_tokens": 5816362.0,
      "step": 2590
    },
    {
      "entropy": 5.620948266983032,
      "epoch": 2.8058409951325043,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045971286625588796,
      "loss": 5.3753,
      "mean_token_accuracy": 0.18987466841936113,
      "num_tokens": 5827518.0,
      "step": 2595
    },
    {
      "entropy": 5.806012153625488,
      "epoch": 2.8112493239588967,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004594676156301383,
      "loss": 5.5029,
      "mean_token_accuracy": 0.1827402338385582,
      "num_tokens": 5838455.0,
      "step": 2600
    },
    {
      "entropy": 5.71817798614502,
      "epoch": 2.816657652785289,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045922169464864725,
      "loss": 5.389,
      "mean_token_accuracy": 0.19094510227441788,
      "num_tokens": 5850065.0,
      "step": 2605
    },
    {
      "entropy": 5.711734294891357,
      "epoch": 2.822065981611682,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000458975104205092,
      "loss": 5.5202,
      "mean_token_accuracy": 0.18257656395435334,
      "num_tokens": 5860444.0,
      "step": 2610
    },
    {
      "entropy": 5.784728574752807,
      "epoch": 2.827474310438075,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00045872784519558304,
      "loss": 5.489,
      "mean_token_accuracy": 0.18090750128030778,
      "num_tokens": 5873984.0,
      "step": 2615
    },
    {
      "entropy": 5.746411752700806,
      "epoch": 2.8328826392644673,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004584799185186599,
      "loss": 5.5212,
      "mean_token_accuracy": 0.18793384730815887,
      "num_tokens": 5885807.0,
      "step": 2620
    },
    {
      "entropy": 5.704473447799683,
      "epoch": 2.8382909680908597,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045823132507528903,
      "loss": 5.4024,
      "mean_token_accuracy": 0.18671886324882508,
      "num_tokens": 5897122.0,
      "step": 2625
    },
    {
      "entropy": 5.63995680809021,
      "epoch": 2.8436992969172525,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004579820657688593,
      "loss": 5.3991,
      "mean_token_accuracy": 0.18967421501874923,
      "num_tokens": 5909189.0,
      "step": 2630
    },
    {
      "entropy": 5.745748233795166,
      "epoch": 2.8491076257436454,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004577321415051798,
      "loss": 5.4968,
      "mean_token_accuracy": 0.1795533388853073,
      "num_tokens": 5920135.0,
      "step": 2635
    },
    {
      "entropy": 5.72193169593811,
      "epoch": 2.854515954570038,
      "grad_norm": 1.0,
      "learning_rate": 0.00045748155319247564,
      "loss": 5.4737,
      "mean_token_accuracy": 0.187947179377079,
      "num_tokens": 5932279.0,
      "step": 2640
    },
    {
      "entropy": 5.602014493942261,
      "epoch": 2.8599242833964302,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004572303017413856,
      "loss": 5.4078,
      "mean_token_accuracy": 0.18369463682174683,
      "num_tokens": 5942082.0,
      "step": 2645
    },
    {
      "entropy": 5.815732002258301,
      "epoch": 2.865332612222823,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004569783880649578,
      "loss": 5.4523,
      "mean_token_accuracy": 0.1847761943936348,
      "num_tokens": 5952856.0,
      "step": 2650
    },
    {
      "entropy": 5.584131813049316,
      "epoch": 2.870740941049216,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004567258130786474,
      "loss": 5.3521,
      "mean_token_accuracy": 0.1931051105260849,
      "num_tokens": 5963937.0,
      "step": 2655
    },
    {
      "entropy": 5.698371505737304,
      "epoch": 2.8761492698756084,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045647257770031215,
      "loss": 5.4446,
      "mean_token_accuracy": 0.1882467731833458,
      "num_tokens": 5974852.0,
      "step": 2660
    },
    {
      "entropy": 5.723794889450073,
      "epoch": 2.8815575987020012,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045621868285021023,
      "loss": 5.4307,
      "mean_token_accuracy": 0.1901254966855049,
      "num_tokens": 5985616.0,
      "step": 2665
    },
    {
      "entropy": 5.709406042098999,
      "epoch": 2.8869659275283936,
      "grad_norm": 1.25,
      "learning_rate": 0.00045596412945099615,
      "loss": 5.3944,
      "mean_token_accuracy": 0.19327322244644166,
      "num_tokens": 5997199.0,
      "step": 2670
    },
    {
      "entropy": 5.7356147289276125,
      "epoch": 2.8923742563547865,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045570891842771727,
      "loss": 5.4415,
      "mean_token_accuracy": 0.18972726613283158,
      "num_tokens": 6008654.0,
      "step": 2675
    },
    {
      "entropy": 5.670080661773682,
      "epoch": 2.897782585181179,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045545305070781135,
      "loss": 5.5062,
      "mean_token_accuracy": 0.18060076087713242,
      "num_tokens": 6020662.0,
      "step": 2680
    },
    {
      "entropy": 5.729869079589844,
      "epoch": 2.903190914007572,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045519652722110206,
      "loss": 5.3768,
      "mean_token_accuracy": 0.205011285841465,
      "num_tokens": 6031793.0,
      "step": 2685
    },
    {
      "entropy": 5.637052536010742,
      "epoch": 2.908599242833964,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045493934889979655,
      "loss": 5.3676,
      "mean_token_accuracy": 0.19111330062150955,
      "num_tokens": 6042181.0,
      "step": 2690
    },
    {
      "entropy": 5.5539429664611815,
      "epoch": 2.914007571660357,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00045468151667848116,
      "loss": 5.3319,
      "mean_token_accuracy": 0.19787560701370238,
      "num_tokens": 6052774.0,
      "step": 2695
    },
    {
      "entropy": 5.761758470535279,
      "epoch": 2.9194159004867495,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000454423031494119,
      "loss": 5.3492,
      "mean_token_accuracy": 0.20007340908050536,
      "num_tokens": 6063273.0,
      "step": 2700
    },
    {
      "entropy": 5.633448791503906,
      "epoch": 2.9248242293131423,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004541638942860458,
      "loss": 5.3516,
      "mean_token_accuracy": 0.1881837248802185,
      "num_tokens": 6075121.0,
      "step": 2705
    },
    {
      "entropy": 5.552972984313965,
      "epoch": 2.9302325581395348,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004539041059959667,
      "loss": 5.3285,
      "mean_token_accuracy": 0.19073624461889266,
      "num_tokens": 6085666.0,
      "step": 2710
    },
    {
      "entropy": 5.6588763236999515,
      "epoch": 2.9356408869659276,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045364366756795295,
      "loss": 5.3503,
      "mean_token_accuracy": 0.19836849421262742,
      "num_tokens": 6096619.0,
      "step": 2715
    },
    {
      "entropy": 5.739606475830078,
      "epoch": 2.94104921579232,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00045338257994843845,
      "loss": 5.5105,
      "mean_token_accuracy": 0.1829600602388382,
      "num_tokens": 6108800.0,
      "step": 2720
    },
    {
      "entropy": 5.694767618179322,
      "epoch": 2.946457544618713,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004531208440862162,
      "loss": 5.4299,
      "mean_token_accuracy": 0.184509801864624,
      "num_tokens": 6119789.0,
      "step": 2725
    },
    {
      "entropy": 5.6493161678314205,
      "epoch": 2.9518658734451053,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045285846093243485,
      "loss": 5.4109,
      "mean_token_accuracy": 0.18593295961618422,
      "num_tokens": 6131786.0,
      "step": 2730
    },
    {
      "entropy": 5.778415393829346,
      "epoch": 2.957274202271498,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045259543144059557,
      "loss": 5.4955,
      "mean_token_accuracy": 0.18639069199562072,
      "num_tokens": 6143012.0,
      "step": 2735
    },
    {
      "entropy": 5.680844163894653,
      "epoch": 2.9626825310978906,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004523317565665479,
      "loss": 5.3744,
      "mean_token_accuracy": 0.19005105644464493,
      "num_tokens": 6153972.0,
      "step": 2740
    },
    {
      "entropy": 5.6828240871429445,
      "epoch": 2.9680908599242835,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00045206743726848716,
      "loss": 5.5606,
      "mean_token_accuracy": 0.18326651751995088,
      "num_tokens": 6166202.0,
      "step": 2745
    },
    {
      "entropy": 5.646218872070312,
      "epoch": 2.973499188750676,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004518024745069502,
      "loss": 5.2738,
      "mean_token_accuracy": 0.19796675592660903,
      "num_tokens": 6177253.0,
      "step": 2750
    },
    {
      "entropy": 5.706516933441162,
      "epoch": 2.9789075175770687,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004515368692448123,
      "loss": 5.5423,
      "mean_token_accuracy": 0.17865623086690902,
      "num_tokens": 6188857.0,
      "step": 2755
    },
    {
      "entropy": 5.7656073570251465,
      "epoch": 2.9843158464034616,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045127062244728353,
      "loss": 5.495,
      "mean_token_accuracy": 0.17895126789808274,
      "num_tokens": 6199945.0,
      "step": 2760
    },
    {
      "entropy": 5.680928325653076,
      "epoch": 2.989724175229854,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004510037350819055,
      "loss": 5.4886,
      "mean_token_accuracy": 0.18967449814081191,
      "num_tokens": 6209866.0,
      "step": 2765
    },
    {
      "entropy": 5.743233966827392,
      "epoch": 2.9951325040562464,
      "grad_norm": 0.875,
      "learning_rate": 0.00045073620811854744,
      "loss": 5.4464,
      "mean_token_accuracy": 0.1788336917757988,
      "num_tokens": 6221636.0,
      "step": 2770
    },
    {
      "entropy": 5.629910839928521,
      "epoch": 3.0,
      "grad_norm": 2.328125,
      "learning_rate": 0.00045046804252940294,
      "loss": 5.3297,
      "mean_token_accuracy": 0.18830747074551052,
      "num_tokens": 6230667.0,
      "step": 2775
    },
    {
      "entropy": 5.69707932472229,
      "epoch": 3.005408328826393,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045019923928898644,
      "loss": 5.118,
      "mean_token_accuracy": 0.21316049247980118,
      "num_tokens": 6242242.0,
      "step": 2780
    },
    {
      "entropy": 5.7481467723846436,
      "epoch": 3.0108166576527853,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004499297993741294,
      "loss": 5.1797,
      "mean_token_accuracy": 0.20156483650207518,
      "num_tokens": 6253337.0,
      "step": 2785
    },
    {
      "entropy": 5.593620777130127,
      "epoch": 3.016224986479178,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004496597237639773,
      "loss": 5.0569,
      "mean_token_accuracy": 0.20687109231948853,
      "num_tokens": 6264582.0,
      "step": 2790
    },
    {
      "entropy": 5.660362005233765,
      "epoch": 3.0216333153055706,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044938901343998536,
      "loss": 5.0992,
      "mean_token_accuracy": 0.21259990483522415,
      "num_tokens": 6276172.0,
      "step": 2795
    },
    {
      "entropy": 5.587935733795166,
      "epoch": 3.0270416441319634,
      "grad_norm": 1.125,
      "learning_rate": 0.0004491176693859156,
      "loss": 4.9704,
      "mean_token_accuracy": 0.22129097431898118,
      "num_tokens": 6286038.0,
      "step": 2800
    },
    {
      "entropy": 5.540070295333862,
      "epoch": 3.032449972958356,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004488456925878329,
      "loss": 5.0489,
      "mean_token_accuracy": 0.21043426394462586,
      "num_tokens": 6296546.0,
      "step": 2805
    },
    {
      "entropy": 5.669938659667968,
      "epoch": 3.0378583017847487,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044857308403410163,
      "loss": 5.1373,
      "mean_token_accuracy": 0.21281687468290328,
      "num_tokens": 6308001.0,
      "step": 2810
    },
    {
      "entropy": 5.629265689849854,
      "epoch": 3.043266630611141,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00044829984471538183,
      "loss": 5.1439,
      "mean_token_accuracy": 0.20406004339456557,
      "num_tokens": 6320896.0,
      "step": 2815
    },
    {
      "entropy": 5.689284133911133,
      "epoch": 3.048674959437534,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004480259756246258,
      "loss": 5.2108,
      "mean_token_accuracy": 0.2020654261112213,
      "num_tokens": 6332710.0,
      "step": 2820
    },
    {
      "entropy": 5.629602384567261,
      "epoch": 3.0540832882639264,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044775147775707465,
      "loss": 5.0681,
      "mean_token_accuracy": 0.21166091561317443,
      "num_tokens": 6343563.0,
      "step": 2825
    },
    {
      "entropy": 5.615605401992798,
      "epoch": 3.0594916170903192,
      "grad_norm": 1.125,
      "learning_rate": 0.000447476352110254,
      "loss": 5.1731,
      "mean_token_accuracy": 0.20783162117004395,
      "num_tokens": 6354064.0,
      "step": 2830
    },
    {
      "entropy": 5.519037961959839,
      "epoch": 3.0648999459167117,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004472005996839712,
      "loss": 5.1062,
      "mean_token_accuracy": 0.20571049004793168,
      "num_tokens": 6366551.0,
      "step": 2835
    },
    {
      "entropy": 5.60793662071228,
      "epoch": 3.0703082747431045,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004469242214803112,
      "loss": 5.0352,
      "mean_token_accuracy": 0.212779301404953,
      "num_tokens": 6377086.0,
      "step": 2840
    },
    {
      "entropy": 5.6298433303833,
      "epoch": 3.075716603569497,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004466472185036331,
      "loss": 5.1482,
      "mean_token_accuracy": 0.20868430733680726,
      "num_tokens": 6388795.0,
      "step": 2845
    },
    {
      "entropy": 5.626946496963501,
      "epoch": 3.08112493239589,
      "grad_norm": 1.125,
      "learning_rate": 0.0004463695917605663,
      "loss": 5.1296,
      "mean_token_accuracy": 0.2113844335079193,
      "num_tokens": 6400174.0,
      "step": 2850
    },
    {
      "entropy": 5.570234155654907,
      "epoch": 3.086533261222282,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000446091342260007,
      "loss": 5.0936,
      "mean_token_accuracy": 0.20676784962415695,
      "num_tokens": 6411297.0,
      "step": 2855
    },
    {
      "entropy": 5.585538625717163,
      "epoch": 3.091941590048675,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004458124710131145,
      "loss": 5.1134,
      "mean_token_accuracy": 0.20946162492036818,
      "num_tokens": 6423487.0,
      "step": 2860
    },
    {
      "entropy": 5.724478244781494,
      "epoch": 3.0973499188750675,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044553297903330766,
      "loss": 5.292,
      "mean_token_accuracy": 0.1967707097530365,
      "num_tokens": 6435460.0,
      "step": 2865
    },
    {
      "entropy": 5.722478771209717,
      "epoch": 3.1027582477014604,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044525286733626085,
      "loss": 5.2367,
      "mean_token_accuracy": 0.20661808997392656,
      "num_tokens": 6445664.0,
      "step": 2870
    },
    {
      "entropy": 5.563872766494751,
      "epoch": 3.1081665765278528,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004449721369399008,
      "loss": 5.0886,
      "mean_token_accuracy": 0.20634377598762513,
      "num_tokens": 6457182.0,
      "step": 2875
    },
    {
      "entropy": 5.501190280914306,
      "epoch": 3.1135749053542456,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044469078886440227,
      "loss": 4.9532,
      "mean_token_accuracy": 0.22265869826078416,
      "num_tokens": 6468378.0,
      "step": 2880
    },
    {
      "entropy": 5.541498279571533,
      "epoch": 3.118983234180638,
      "grad_norm": 1.125,
      "learning_rate": 0.00044440882413218484,
      "loss": 5.187,
      "mean_token_accuracy": 0.20512945502996444,
      "num_tokens": 6479625.0,
      "step": 2885
    },
    {
      "entropy": 5.571180772781372,
      "epoch": 3.124391563007031,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044412624376790905,
      "loss": 5.0993,
      "mean_token_accuracy": 0.20619929283857347,
      "num_tokens": 6490713.0,
      "step": 2890
    },
    {
      "entropy": 5.489297723770141,
      "epoch": 3.1297998918334233,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004438430487984726,
      "loss": 5.1367,
      "mean_token_accuracy": 0.1967697337269783,
      "num_tokens": 6502161.0,
      "step": 2895
    },
    {
      "entropy": 5.6030020236969,
      "epoch": 3.135208220659816,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004435592402530066,
      "loss": 5.0473,
      "mean_token_accuracy": 0.21011342406272887,
      "num_tokens": 6513639.0,
      "step": 2900
    },
    {
      "entropy": 5.564647388458252,
      "epoch": 3.1406165494862086,
      "grad_norm": 1.09375,
      "learning_rate": 0.000443274819162872,
      "loss": 5.0812,
      "mean_token_accuracy": 0.20863993614912033,
      "num_tokens": 6524893.0,
      "step": 2905
    },
    {
      "entropy": 5.48593111038208,
      "epoch": 3.1460248783126015,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004429897865616557,
      "loss": 5.0748,
      "mean_token_accuracy": 0.2106839030981064,
      "num_tokens": 6535797.0,
      "step": 2910
    },
    {
      "entropy": 5.594881916046143,
      "epoch": 3.151433207138994,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004427041434851668,
      "loss": 5.167,
      "mean_token_accuracy": 0.20228979587554932,
      "num_tokens": 6547117.0,
      "step": 2915
    },
    {
      "entropy": 5.646280956268311,
      "epoch": 3.1568415359653867,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044241789097143293,
      "loss": 5.1957,
      "mean_token_accuracy": 0.20873761922121048,
      "num_tokens": 6558771.0,
      "step": 2920
    },
    {
      "entropy": 5.559993886947632,
      "epoch": 3.162249864791779,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044213103006069635,
      "loss": 5.1634,
      "mean_token_accuracy": 0.20020761936903,
      "num_tokens": 6569922.0,
      "step": 2925
    },
    {
      "entropy": 5.6437593460083,
      "epoch": 3.167658193618172,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044184356179541035,
      "loss": 5.1619,
      "mean_token_accuracy": 0.20496343672275544,
      "num_tokens": 6580561.0,
      "step": 2930
    },
    {
      "entropy": 5.561740875244141,
      "epoch": 3.1730665224445644,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004415554872202352,
      "loss": 5.0106,
      "mean_token_accuracy": 0.20989638566970825,
      "num_tokens": 6591636.0,
      "step": 2935
    },
    {
      "entropy": 5.564278173446655,
      "epoch": 3.1784748512709573,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044126680738203466,
      "loss": 5.2208,
      "mean_token_accuracy": 0.20207992047071457,
      "num_tokens": 6603736.0,
      "step": 2940
    },
    {
      "entropy": 5.697408962249756,
      "epoch": 3.1838831800973497,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004409775233298718,
      "loss": 5.1527,
      "mean_token_accuracy": 0.2086891621351242,
      "num_tokens": 6615415.0,
      "step": 2945
    },
    {
      "entropy": 5.55486102104187,
      "epoch": 3.1892915089237426,
      "grad_norm": 1.125,
      "learning_rate": 0.0004406876361150056,
      "loss": 5.1808,
      "mean_token_accuracy": 0.20389774590730667,
      "num_tokens": 6625970.0,
      "step": 2950
    },
    {
      "entropy": 5.594521570205688,
      "epoch": 3.194699837750135,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004403971467908869,
      "loss": 5.1384,
      "mean_token_accuracy": 0.20423722565174102,
      "num_tokens": 6636326.0,
      "step": 2955
    },
    {
      "entropy": 5.557396173477173,
      "epoch": 3.200108166576528,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004401060564131545,
      "loss": 5.1705,
      "mean_token_accuracy": 0.20355864316225053,
      "num_tokens": 6646825.0,
      "step": 2960
    },
    {
      "entropy": 5.588989734649658,
      "epoch": 3.2055164954029207,
      "grad_norm": 1.25,
      "learning_rate": 0.0004398143660396315,
      "loss": 5.1703,
      "mean_token_accuracy": 0.20732996463775635,
      "num_tokens": 6657748.0,
      "step": 2965
    },
    {
      "entropy": 5.5528037548065186,
      "epoch": 3.210924824229313,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004395220767303215,
      "loss": 5.0564,
      "mean_token_accuracy": 0.21476534754037857,
      "num_tokens": 6668422.0,
      "step": 2970
    },
    {
      "entropy": 5.615345191955567,
      "epoch": 3.2163331530557056,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004392291895474045,
      "loss": 5.1522,
      "mean_token_accuracy": 0.2010224476456642,
      "num_tokens": 6678898.0,
      "step": 2975
    },
    {
      "entropy": 5.608639144897461,
      "epoch": 3.2217414818820984,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004389357055552331,
      "loss": 5.1839,
      "mean_token_accuracy": 0.21051690131425857,
      "num_tokens": 6688691.0,
      "step": 2980
    },
    {
      "entropy": 5.552885484695435,
      "epoch": 3.2271498107084913,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043864162582032897,
      "loss": 5.075,
      "mean_token_accuracy": 0.21406530737876892,
      "num_tokens": 6700279.0,
      "step": 2985
    },
    {
      "entropy": 5.483924531936646,
      "epoch": 3.2325581395348837,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004383469514113784,
      "loss": 5.0118,
      "mean_token_accuracy": 0.2145751088857651,
      "num_tokens": 6710440.0,
      "step": 2990
    },
    {
      "entropy": 5.53403868675232,
      "epoch": 3.2379664683612766,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004380516833992289,
      "loss": 5.1962,
      "mean_token_accuracy": 0.1990383207798004,
      "num_tokens": 6722347.0,
      "step": 2995
    },
    {
      "entropy": 5.590388536453247,
      "epoch": 3.243374797187669,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000437755822856885,
      "loss": 5.2089,
      "mean_token_accuracy": 0.20684486180543898,
      "num_tokens": 6732936.0,
      "step": 3000
    },
    {
      "epoch": 3.243374797187669,
      "eval_entropy": 5.37777090397748,
      "eval_loss": 5.91184663772583,
      "eval_mean_token_accuracy": 0.17475627675991168,
      "eval_num_tokens": 6732936.0,
      "eval_runtime": 1.9364,
      "eval_samples_per_second": 1815.19,
      "eval_steps_per_second": 227.222,
      "step": 3000
    },
    {
      "entropy": 5.573069620132446,
      "epoch": 3.248783126014062,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004374593708595049,
      "loss": 5.2147,
      "mean_token_accuracy": 0.2031557783484459,
      "num_tokens": 6744552.0,
      "step": 3005
    },
    {
      "entropy": 5.652343273162842,
      "epoch": 3.2541914548404542,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043716232848439555,
      "loss": 5.1481,
      "mean_token_accuracy": 0.20604717880487441,
      "num_tokens": 6755749.0,
      "step": 3010
    },
    {
      "entropy": 5.569940423965454,
      "epoch": 3.259599783666847,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043686469681100985,
      "loss": 5.0899,
      "mean_token_accuracy": 0.20888536125421525,
      "num_tokens": 6766119.0,
      "step": 3015
    },
    {
      "entropy": 5.460270977020263,
      "epoch": 3.2650081124932395,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043656647692094186,
      "loss": 5.1313,
      "mean_token_accuracy": 0.20296620875597,
      "num_tokens": 6776491.0,
      "step": 3020
    },
    {
      "entropy": 5.69976396560669,
      "epoch": 3.2704164413196324,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004362676698979233,
      "loss": 5.239,
      "mean_token_accuracy": 0.20118464827537536,
      "num_tokens": 6788795.0,
      "step": 3025
    },
    {
      "entropy": 5.628712368011475,
      "epoch": 3.275824770146025,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043596827682781974,
      "loss": 5.233,
      "mean_token_accuracy": 0.1996966302394867,
      "num_tokens": 6799706.0,
      "step": 3030
    },
    {
      "entropy": 5.597583246231079,
      "epoch": 3.2812330989724177,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004356682987986262,
      "loss": 5.1885,
      "mean_token_accuracy": 0.20046019405126572,
      "num_tokens": 6812454.0,
      "step": 3035
    },
    {
      "entropy": 5.620301389694214,
      "epoch": 3.28664142779881,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004353677369004635,
      "loss": 5.1281,
      "mean_token_accuracy": 0.20747572928667068,
      "num_tokens": 6823950.0,
      "step": 3040
    },
    {
      "entropy": 5.635411548614502,
      "epoch": 3.292049756625203,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043506659222557426,
      "loss": 5.3024,
      "mean_token_accuracy": 0.19502145648002625,
      "num_tokens": 6835028.0,
      "step": 3045
    },
    {
      "entropy": 5.581868124008179,
      "epoch": 3.2974580854515954,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000434764865868319,
      "loss": 5.1831,
      "mean_token_accuracy": 0.19999626278877258,
      "num_tokens": 6846316.0,
      "step": 3050
    },
    {
      "entropy": 5.527811479568482,
      "epoch": 3.302866414277988,
      "grad_norm": 1.21875,
      "learning_rate": 0.000434462558925172,
      "loss": 5.1175,
      "mean_token_accuracy": 0.20996025204658508,
      "num_tokens": 6856463.0,
      "step": 3055
    },
    {
      "entropy": 5.623858213424683,
      "epoch": 3.3082747431043806,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00043415967249471734,
      "loss": 5.1497,
      "mean_token_accuracy": 0.20746323466300964,
      "num_tokens": 6867869.0,
      "step": 3060
    },
    {
      "entropy": 5.582069349288941,
      "epoch": 3.3136830719307735,
      "grad_norm": 1.09375,
      "learning_rate": 0.000433856207677645,
      "loss": 5.1261,
      "mean_token_accuracy": 0.20743919014930726,
      "num_tokens": 6878545.0,
      "step": 3065
    },
    {
      "entropy": 5.489497947692871,
      "epoch": 3.319091400757166,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000433552165576747,
      "loss": 5.1641,
      "mean_token_accuracy": 0.20064468085765838,
      "num_tokens": 6889399.0,
      "step": 3070
    },
    {
      "entropy": 5.624862194061279,
      "epoch": 3.3244997295835588,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043324754729691275,
      "loss": 5.1596,
      "mean_token_accuracy": 0.2113773137331009,
      "num_tokens": 6900428.0,
      "step": 3075
    },
    {
      "entropy": 5.69494662284851,
      "epoch": 3.329908058409951,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043294235394512594,
      "loss": 5.2702,
      "mean_token_accuracy": 0.20045885890722276,
      "num_tokens": 6911112.0,
      "step": 3080
    },
    {
      "entropy": 5.521151447296143,
      "epoch": 3.335316387236344,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004326365866304599,
      "loss": 5.0236,
      "mean_token_accuracy": 0.21755116283893586,
      "num_tokens": 6922383.0,
      "step": 3085
    },
    {
      "entropy": 5.547823572158814,
      "epoch": 3.3407247160627365,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004323302464640738,
      "loss": 5.222,
      "mean_token_accuracy": 0.19735278338193893,
      "num_tokens": 6935153.0,
      "step": 3090
    },
    {
      "entropy": 5.670290184020996,
      "epoch": 3.3461330448891293,
      "grad_norm": 1.03125,
      "learning_rate": 0.00043202333455920843,
      "loss": 5.2639,
      "mean_token_accuracy": 0.20184647887945176,
      "num_tokens": 6946814.0,
      "step": 3095
    },
    {
      "entropy": 5.546437168121338,
      "epoch": 3.3515413737155217,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004317158520311824,
      "loss": 5.1129,
      "mean_token_accuracy": 0.2076606571674347,
      "num_tokens": 6957230.0,
      "step": 3100
    },
    {
      "entropy": 5.44480447769165,
      "epoch": 3.3569497025419146,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004314077999973879,
      "loss": 5.1318,
      "mean_token_accuracy": 0.19855479449033736,
      "num_tokens": 6968060.0,
      "step": 3105
    },
    {
      "entropy": 5.568084287643432,
      "epoch": 3.362358031368307,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004310991795772869,
      "loss": 5.0431,
      "mean_token_accuracy": 0.21078354120254517,
      "num_tokens": 6977782.0,
      "step": 3110
    },
    {
      "entropy": 5.608031511306763,
      "epoch": 3.3677663601947,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004307899918924065,
      "loss": 5.1228,
      "mean_token_accuracy": 0.20318224132061005,
      "num_tokens": 6988252.0,
      "step": 3115
    },
    {
      "entropy": 5.598913288116455,
      "epoch": 3.3731746890210923,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004304802380663356,
      "loss": 5.1324,
      "mean_token_accuracy": 0.2066564679145813,
      "num_tokens": 6998523.0,
      "step": 3120
    },
    {
      "entropy": 5.519530439376831,
      "epoch": 3.378583017847485,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043016991922472015,
      "loss": 5.1628,
      "mean_token_accuracy": 0.2022525265812874,
      "num_tokens": 7010058.0,
      "step": 3125
    },
    {
      "entropy": 5.593735790252685,
      "epoch": 3.383991346673878,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042985903649525977,
      "loss": 5.2746,
      "mean_token_accuracy": 0.19244320541620255,
      "num_tokens": 7021289.0,
      "step": 3130
    },
    {
      "entropy": 5.713921213150025,
      "epoch": 3.3893996755002704,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042954759100770275,
      "loss": 5.3005,
      "mean_token_accuracy": 0.1927209183573723,
      "num_tokens": 7032649.0,
      "step": 3135
    },
    {
      "entropy": 5.635376024246216,
      "epoch": 3.394808004326663,
      "grad_norm": 1.234375,
      "learning_rate": 0.00042923558389384303,
      "loss": 5.1584,
      "mean_token_accuracy": 0.2046155095100403,
      "num_tokens": 7044120.0,
      "step": 3140
    },
    {
      "entropy": 5.5387166976928714,
      "epoch": 3.4002163331530557,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004289230162875149,
      "loss": 5.1575,
      "mean_token_accuracy": 0.20784637182950974,
      "num_tokens": 7054526.0,
      "step": 3145
    },
    {
      "entropy": 5.431732416152954,
      "epoch": 3.4056246619794486,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042860988932458987,
      "loss": 5.0371,
      "mean_token_accuracy": 0.21920679807662963,
      "num_tokens": 7064744.0,
      "step": 3150
    },
    {
      "entropy": 5.573424243927002,
      "epoch": 3.411032990805841,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004282962041429721,
      "loss": 5.1564,
      "mean_token_accuracy": 0.20835559219121932,
      "num_tokens": 7076360.0,
      "step": 3155
    },
    {
      "entropy": 5.627830553054809,
      "epoch": 3.4164413196322334,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042798196188259427,
      "loss": 5.1758,
      "mean_token_accuracy": 0.20924926102161406,
      "num_tokens": 7088114.0,
      "step": 3160
    },
    {
      "entropy": 5.613403463363648,
      "epoch": 3.4218496484586263,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004276671636854133,
      "loss": 5.2014,
      "mean_token_accuracy": 0.19810499548912047,
      "num_tokens": 7099125.0,
      "step": 3165
    },
    {
      "entropy": 5.5529709815979,
      "epoch": 3.427257977285019,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042735181069540677,
      "loss": 5.1391,
      "mean_token_accuracy": 0.21497857421636582,
      "num_tokens": 7109875.0,
      "step": 3170
    },
    {
      "entropy": 5.55393533706665,
      "epoch": 3.4326663061114115,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004270359040585681,
      "loss": 5.1845,
      "mean_token_accuracy": 0.20396130681037902,
      "num_tokens": 7121388.0,
      "step": 3175
    },
    {
      "entropy": 5.5741424560546875,
      "epoch": 3.4380746349378044,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042671944492290277,
      "loss": 5.21,
      "mean_token_accuracy": 0.19878134727478028,
      "num_tokens": 7134161.0,
      "step": 3180
    },
    {
      "entropy": 5.624898862838745,
      "epoch": 3.443482963764197,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004264024344384239,
      "loss": 5.2275,
      "mean_token_accuracy": 0.1993679642677307,
      "num_tokens": 7146580.0,
      "step": 3185
    },
    {
      "entropy": 5.573002195358276,
      "epoch": 3.4488912925905897,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042608487375714847,
      "loss": 5.1357,
      "mean_token_accuracy": 0.19975454062223436,
      "num_tokens": 7156969.0,
      "step": 3190
    },
    {
      "entropy": 5.560785627365112,
      "epoch": 3.454299621416982,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042576676403309265,
      "loss": 5.0973,
      "mean_token_accuracy": 0.21098205000162124,
      "num_tokens": 7168687.0,
      "step": 3195
    },
    {
      "entropy": 5.597518825531006,
      "epoch": 3.459707950243375,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004254481064222679,
      "loss": 5.1821,
      "mean_token_accuracy": 0.20848626345396043,
      "num_tokens": 7179996.0,
      "step": 3200
    },
    {
      "entropy": 5.607846450805664,
      "epoch": 3.4651162790697674,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042512890208267677,
      "loss": 5.213,
      "mean_token_accuracy": 0.19970757514238358,
      "num_tokens": 7191214.0,
      "step": 3205
    },
    {
      "entropy": 5.581211709976197,
      "epoch": 3.4705246078961602,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042480915217430853,
      "loss": 5.2306,
      "mean_token_accuracy": 0.20193232595920563,
      "num_tokens": 7203491.0,
      "step": 3210
    },
    {
      "entropy": 5.650781297683716,
      "epoch": 3.4759329367225527,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042448885785913507,
      "loss": 5.1573,
      "mean_token_accuracy": 0.20115691423416138,
      "num_tokens": 7215018.0,
      "step": 3215
    },
    {
      "entropy": 5.562850618362427,
      "epoch": 3.4813412655489455,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042416802030110677,
      "loss": 5.1011,
      "mean_token_accuracy": 0.2093229651451111,
      "num_tokens": 7226209.0,
      "step": 3220
    },
    {
      "entropy": 5.552242565155029,
      "epoch": 3.486749594375338,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042384664066614787,
      "loss": 5.1159,
      "mean_token_accuracy": 0.2053811177611351,
      "num_tokens": 7236505.0,
      "step": 3225
    },
    {
      "entropy": 5.487067079544067,
      "epoch": 3.492157923201731,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004235247201221528,
      "loss": 5.0375,
      "mean_token_accuracy": 0.21327000260353088,
      "num_tokens": 7247414.0,
      "step": 3230
    },
    {
      "entropy": 5.582530832290649,
      "epoch": 3.497566252028123,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004232022598389816,
      "loss": 5.2253,
      "mean_token_accuracy": 0.2056236296892166,
      "num_tokens": 7258666.0,
      "step": 3235
    },
    {
      "entropy": 5.610252571105957,
      "epoch": 3.502974580854516,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004228792609884557,
      "loss": 5.1518,
      "mean_token_accuracy": 0.20571167320013045,
      "num_tokens": 7270201.0,
      "step": 3240
    },
    {
      "entropy": 5.6072530269622805,
      "epoch": 3.5083829096809085,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042255572474435364,
      "loss": 5.2132,
      "mean_token_accuracy": 0.20459086447954178,
      "num_tokens": 7282254.0,
      "step": 3245
    },
    {
      "entropy": 5.589420318603516,
      "epoch": 3.5137912385073014,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004222316522824069,
      "loss": 5.2566,
      "mean_token_accuracy": 0.19173479825258255,
      "num_tokens": 7294579.0,
      "step": 3250
    },
    {
      "entropy": 5.562285947799682,
      "epoch": 3.5191995673336938,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004219070447802955,
      "loss": 5.2104,
      "mean_token_accuracy": 0.19856292456388475,
      "num_tokens": 7305822.0,
      "step": 3255
    },
    {
      "entropy": 5.595698451995849,
      "epoch": 3.5246078961600866,
      "grad_norm": 1.046875,
      "learning_rate": 0.00042158190341764384,
      "loss": 5.2233,
      "mean_token_accuracy": 0.2017119914293289,
      "num_tokens": 7317543.0,
      "step": 3260
    },
    {
      "entropy": 5.565688610076904,
      "epoch": 3.530016224986479,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042125622937601645,
      "loss": 5.1714,
      "mean_token_accuracy": 0.20284906327724456,
      "num_tokens": 7327804.0,
      "step": 3265
    },
    {
      "entropy": 5.535323143005371,
      "epoch": 3.535424553812872,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004209300238389135,
      "loss": 5.1222,
      "mean_token_accuracy": 0.20388976186513902,
      "num_tokens": 7337045.0,
      "step": 3270
    },
    {
      "entropy": 5.550784778594971,
      "epoch": 3.5408328826392643,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004206032879917667,
      "loss": 5.2015,
      "mean_token_accuracy": 0.21165899336338043,
      "num_tokens": 7348184.0,
      "step": 3275
    },
    {
      "entropy": 5.5675712585449215,
      "epoch": 3.546241211465657,
      "grad_norm": 1.25,
      "learning_rate": 0.0004202760230219348,
      "loss": 5.1187,
      "mean_token_accuracy": 0.20928198099136353,
      "num_tokens": 7359571.0,
      "step": 3280
    },
    {
      "entropy": 5.499993753433228,
      "epoch": 3.5516495402920496,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041994823011869955,
      "loss": 5.1258,
      "mean_token_accuracy": 0.20613025724887848,
      "num_tokens": 7370429.0,
      "step": 3285
    },
    {
      "entropy": 5.59150800704956,
      "epoch": 3.5570578691184425,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041961991047326116,
      "loss": 5.0892,
      "mean_token_accuracy": 0.2082501992583275,
      "num_tokens": 7381569.0,
      "step": 3290
    },
    {
      "entropy": 5.499939250946045,
      "epoch": 3.5624661979448353,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004192910652787338,
      "loss": 5.0841,
      "mean_token_accuracy": 0.20715413987636566,
      "num_tokens": 7392746.0,
      "step": 3295
    },
    {
      "entropy": 5.549461317062378,
      "epoch": 3.5678745267712277,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004189616957301419,
      "loss": 5.1891,
      "mean_token_accuracy": 0.19182858914136885,
      "num_tokens": 7405520.0,
      "step": 3300
    },
    {
      "entropy": 5.501055097579956,
      "epoch": 3.57328285559762,
      "grad_norm": 1.125,
      "learning_rate": 0.0004186318030244149,
      "loss": 5.1565,
      "mean_token_accuracy": 0.2044055238366127,
      "num_tokens": 7417788.0,
      "step": 3305
    },
    {
      "entropy": 5.576556062698364,
      "epoch": 3.578691184424013,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00041830138836038383,
      "loss": 5.2256,
      "mean_token_accuracy": 0.20100464969873427,
      "num_tokens": 7428912.0,
      "step": 3310
    },
    {
      "entropy": 5.583335733413696,
      "epoch": 3.584099513250406,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004179704529387763,
      "loss": 5.1013,
      "mean_token_accuracy": 0.21064824014902114,
      "num_tokens": 7439797.0,
      "step": 3315
    },
    {
      "entropy": 5.462785291671753,
      "epoch": 3.5895078420767983,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004176389979622125,
      "loss": 5.0936,
      "mean_token_accuracy": 0.20896225422620773,
      "num_tokens": 7451755.0,
      "step": 3320
    },
    {
      "entropy": 5.588175821304321,
      "epoch": 3.5949161709031907,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004173070246352003,
      "loss": 5.1926,
      "mean_token_accuracy": 0.20159001499414445,
      "num_tokens": 7463393.0,
      "step": 3325
    },
    {
      "entropy": 5.448696327209473,
      "epoch": 3.6003244997295836,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041697453416413184,
      "loss": 5.1384,
      "mean_token_accuracy": 0.20230467319488527,
      "num_tokens": 7474404.0,
      "step": 3330
    },
    {
      "entropy": 5.559615850448608,
      "epoch": 3.6057328285559764,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000416641527757278,
      "loss": 5.1194,
      "mean_token_accuracy": 0.20770540982484817,
      "num_tokens": 7485495.0,
      "step": 3335
    },
    {
      "entropy": 5.4712920665740965,
      "epoch": 3.611141157382369,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041630800662478514,
      "loss": 5.1036,
      "mean_token_accuracy": 0.20918697565793992,
      "num_tokens": 7497305.0,
      "step": 3340
    },
    {
      "entropy": 5.491168975830078,
      "epoch": 3.6165494862087613,
      "grad_norm": 1.0625,
      "learning_rate": 0.00041597397197866957,
      "loss": 5.0288,
      "mean_token_accuracy": 0.21322695314884185,
      "num_tokens": 7509008.0,
      "step": 3345
    },
    {
      "entropy": 5.470600843429565,
      "epoch": 3.621957815035154,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004156394250328141,
      "loss": 5.1054,
      "mean_token_accuracy": 0.2145393580198288,
      "num_tokens": 7520079.0,
      "step": 3350
    },
    {
      "entropy": 5.501084804534912,
      "epoch": 3.627366143861547,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000415304367002963,
      "loss": 5.1446,
      "mean_token_accuracy": 0.2071783125400543,
      "num_tokens": 7531294.0,
      "step": 3355
    },
    {
      "entropy": 5.548788261413574,
      "epoch": 3.6327744726879394,
      "grad_norm": 1.046875,
      "learning_rate": 0.00041496879910671796,
      "loss": 5.1571,
      "mean_token_accuracy": 0.20829515308141708,
      "num_tokens": 7542508.0,
      "step": 3360
    },
    {
      "entropy": 5.529984998703003,
      "epoch": 3.638182801514332,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004146327225635336,
      "loss": 5.102,
      "mean_token_accuracy": 0.20913522839546203,
      "num_tokens": 7553288.0,
      "step": 3365
    },
    {
      "entropy": 5.604449224472046,
      "epoch": 3.6435911303407247,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004142961385947127,
      "loss": 5.2801,
      "mean_token_accuracy": 0.20308633297681808,
      "num_tokens": 7564532.0,
      "step": 3370
    },
    {
      "entropy": 5.537926387786865,
      "epoch": 3.6489994591671175,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041395904842340215,
      "loss": 5.1105,
      "mean_token_accuracy": 0.21103381514549255,
      "num_tokens": 7575109.0,
      "step": 3375
    },
    {
      "entropy": 5.472426414489746,
      "epoch": 3.65440778799351,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004136214532745887,
      "loss": 5.1106,
      "mean_token_accuracy": 0.20961469113826753,
      "num_tokens": 7585871.0,
      "step": 3380
    },
    {
      "entropy": 5.582272386550903,
      "epoch": 3.6598161168199024,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004132833543750936,
      "loss": 5.1941,
      "mean_token_accuracy": 0.20975491106510163,
      "num_tokens": 7596821.0,
      "step": 3385
    },
    {
      "entropy": 5.451750040054321,
      "epoch": 3.6652244456462952,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041294475295356917,
      "loss": 5.1274,
      "mean_token_accuracy": 0.20204561948776245,
      "num_tokens": 7608101.0,
      "step": 3390
    },
    {
      "entropy": 5.545164632797241,
      "epoch": 3.670632774472688,
      "grad_norm": 1.125,
      "learning_rate": 0.0004126056502404937,
      "loss": 5.1188,
      "mean_token_accuracy": 0.2001086190342903,
      "num_tokens": 7618722.0,
      "step": 3395
    },
    {
      "entropy": 5.506054496765136,
      "epoch": 3.6760411032990805,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004122660474681672,
      "loss": 5.1293,
      "mean_token_accuracy": 0.207930351793766,
      "num_tokens": 7630061.0,
      "step": 3400
    },
    {
      "entropy": 5.536324167251587,
      "epoch": 3.6814494321254734,
      "grad_norm": 1.125,
      "learning_rate": 0.00041192594587070706,
      "loss": 5.1173,
      "mean_token_accuracy": 0.20743480771780015,
      "num_tokens": 7641830.0,
      "step": 3405
    },
    {
      "entropy": 5.484718084335327,
      "epoch": 3.686857760951866,
      "grad_norm": 1.0625,
      "learning_rate": 0.00041158534668404325,
      "loss": 5.1588,
      "mean_token_accuracy": 0.20602032393217087,
      "num_tokens": 7652750.0,
      "step": 3410
    },
    {
      "entropy": 5.5122642993927,
      "epoch": 3.6922660897782587,
      "grad_norm": 1.125,
      "learning_rate": 0.00041124425114591386,
      "loss": 5.065,
      "mean_token_accuracy": 0.21273843944072723,
      "num_tokens": 7663405.0,
      "step": 3415
    },
    {
      "entropy": 5.462223911285401,
      "epoch": 3.697674418604651,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041090266049586107,
      "loss": 5.0796,
      "mean_token_accuracy": 0.19900004267692567,
      "num_tokens": 7673958.0,
      "step": 3420
    },
    {
      "entropy": 5.500379133224487,
      "epoch": 3.703082747431044,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004105605759752258,
      "loss": 5.0774,
      "mean_token_accuracy": 0.2081347331404686,
      "num_tokens": 7685475.0,
      "step": 3425
    },
    {
      "entropy": 5.420930194854736,
      "epoch": 3.7084910762574363,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041021799882714436,
      "loss": 5.0815,
      "mean_token_accuracy": 0.20927014648914338,
      "num_tokens": 7697296.0,
      "step": 3430
    },
    {
      "entropy": 5.539216470718384,
      "epoch": 3.713899405083829,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004098749302965426,
      "loss": 5.2197,
      "mean_token_accuracy": 0.20556779056787491,
      "num_tokens": 7709414.0,
      "step": 3435
    },
    {
      "entropy": 5.562742471694946,
      "epoch": 3.7193077339102216,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004095313716301324,
      "loss": 5.2042,
      "mean_token_accuracy": 0.1960515037178993,
      "num_tokens": 7720965.0,
      "step": 3440
    },
    {
      "entropy": 5.509733247756958,
      "epoch": 3.7247160627366145,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004091873240764069,
      "loss": 5.1567,
      "mean_token_accuracy": 0.19600157886743547,
      "num_tokens": 7733044.0,
      "step": 3445
    },
    {
      "entropy": 5.624178552627564,
      "epoch": 3.730124391563007,
      "grad_norm": 1.296875,
      "learning_rate": 0.00040884278888563566,
      "loss": 5.1338,
      "mean_token_accuracy": 0.2047843962907791,
      "num_tokens": 7743878.0,
      "step": 3450
    },
    {
      "entropy": 5.433482074737549,
      "epoch": 3.7355327203893998,
      "grad_norm": 1.203125,
      "learning_rate": 0.00040849776730986014,
      "loss": 5.1322,
      "mean_token_accuracy": 0.21231569796800615,
      "num_tokens": 7755445.0,
      "step": 3455
    },
    {
      "entropy": 5.584301090240478,
      "epoch": 3.740941049215792,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004081522606028899,
      "loss": 5.1974,
      "mean_token_accuracy": 0.2005298689007759,
      "num_tokens": 7765990.0,
      "step": 3460
    },
    {
      "entropy": 5.445444393157959,
      "epoch": 3.746349378042185,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00040780627002029674,
      "loss": 5.0037,
      "mean_token_accuracy": 0.21728041470050813,
      "num_tokens": 7777064.0,
      "step": 3465
    },
    {
      "entropy": 5.393490314483643,
      "epoch": 3.7517577068685775,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004074597968194116,
      "loss": 5.0414,
      "mean_token_accuracy": 0.2053399369120598,
      "num_tokens": 7788618.0,
      "step": 3470
    },
    {
      "entropy": 5.508790159225464,
      "epoch": 3.7571660356949703,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004071128422593186,
      "loss": 5.1788,
      "mean_token_accuracy": 0.1951671987771988,
      "num_tokens": 7798484.0,
      "step": 3475
    },
    {
      "entropy": 5.535230827331543,
      "epoch": 3.7625743645213627,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004067654076008515,
      "loss": 5.1358,
      "mean_token_accuracy": 0.20426508635282517,
      "num_tokens": 7810270.0,
      "step": 3480
    },
    {
      "entropy": 5.502645969390869,
      "epoch": 3.7679826933477556,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00040641749410658877,
      "loss": 5.2302,
      "mean_token_accuracy": 0.20067734271287918,
      "num_tokens": 7822616.0,
      "step": 3485
    },
    {
      "entropy": 5.5701946258544925,
      "epoch": 3.773391022174148,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004060691030408487,
      "loss": 5.1298,
      "mean_token_accuracy": 0.2053201138973236,
      "num_tokens": 7834933.0,
      "step": 3490
    },
    {
      "entropy": 5.543779182434082,
      "epoch": 3.778799351000541,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040572023566968536,
      "loss": 5.1516,
      "mean_token_accuracy": 0.20760101526975633,
      "num_tokens": 7847686.0,
      "step": 3495
    },
    {
      "entropy": 5.480415725708008,
      "epoch": 3.7842076798269337,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040537089326088345,
      "loss": 5.1359,
      "mean_token_accuracy": 0.20168039649724961,
      "num_tokens": 7859087.0,
      "step": 3500
    },
    {
      "epoch": 3.7842076798269337,
      "eval_entropy": 5.405016609755429,
      "eval_loss": 5.846310615539551,
      "eval_mean_token_accuracy": 0.17764256713403898,
      "eval_num_tokens": 7859087.0,
      "eval_runtime": 1.934,
      "eval_samples_per_second": 1817.514,
      "eval_steps_per_second": 227.512,
      "step": 3500
    }
  ],
  "logging_steps": 5,
  "max_steps": 9240,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1822957332480000.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}