{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.6219578150351541,
  "eval_steps": 500,
  "global_step": 1500,
  "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
    }
  ],
  "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": 775440484761600.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}