isl-latn-100mb-after-ppt-Dp-100mb-ckpt500_seed3407/checkpoint-6000/trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.5931198102016607,
  "eval_steps": 3000,
  "global_step": 6000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 7.151007413864136,
      "epoch": 0.0004942665085013839,
      "grad_norm": 1.078125,
      "learning_rate": 2e-06,
      "loss": 7.279,
      "mean_token_accuracy": 0.08845784291625022,
      "num_tokens": 11289.0,
      "step": 5
    },
    {
      "entropy": 7.142992544174194,
      "epoch": 0.0009885330170027679,
      "grad_norm": 1.1015625,
      "learning_rate": 4.5e-06,
      "loss": 7.2577,
      "mean_token_accuracy": 0.08933138325810433,
      "num_tokens": 21541.0,
      "step": 10
    },
    {
      "entropy": 7.182202959060669,
      "epoch": 0.0014827995255041518,
      "grad_norm": 1.0703125,
      "learning_rate": 7e-06,
      "loss": 7.2317,
      "mean_token_accuracy": 0.0934896469116211,
      "num_tokens": 32257.0,
      "step": 15
    },
    {
      "entropy": 7.244844675064087,
      "epoch": 0.0019770660340055358,
      "grad_norm": 0.9140625,
      "learning_rate": 9.5e-06,
      "loss": 7.2492,
      "mean_token_accuracy": 0.09298909306526185,
      "num_tokens": 44176.0,
      "step": 20
    },
    {
      "entropy": 7.262534332275391,
      "epoch": 0.0024713325425069197,
      "grad_norm": 0.98828125,
      "learning_rate": 1.2e-05,
      "loss": 7.1678,
      "mean_token_accuracy": 0.09061622098088265,
      "num_tokens": 56115.0,
      "step": 25
    },
    {
      "entropy": 7.378157377243042,
      "epoch": 0.0029655990510083037,
      "grad_norm": 0.94921875,
      "learning_rate": 1.4500000000000002e-05,
      "loss": 7.2162,
      "mean_token_accuracy": 0.09287815466523171,
      "num_tokens": 68828.0,
      "step": 30
    },
    {
      "entropy": 7.399503707885742,
      "epoch": 0.0034598655595096876,
      "grad_norm": 0.98828125,
      "learning_rate": 1.7000000000000003e-05,
      "loss": 7.2093,
      "mean_token_accuracy": 0.09124612733721733,
      "num_tokens": 80097.0,
      "step": 35
    },
    {
      "entropy": 7.373995113372803,
      "epoch": 0.0039541320680110716,
      "grad_norm": 0.83203125,
      "learning_rate": 1.95e-05,
      "loss": 7.1903,
      "mean_token_accuracy": 0.09260638207197189,
      "num_tokens": 91510.0,
      "step": 40
    },
    {
      "entropy": 7.330695772171021,
      "epoch": 0.004448398576512456,
      "grad_norm": 0.8046875,
      "learning_rate": 2.2e-05,
      "loss": 7.234,
      "mean_token_accuracy": 0.09660193100571632,
      "num_tokens": 103443.0,
      "step": 45
    },
    {
      "entropy": 7.33756799697876,
      "epoch": 0.0049426650850138395,
      "grad_norm": 0.83984375,
      "learning_rate": 2.4500000000000003e-05,
      "loss": 7.2565,
      "mean_token_accuracy": 0.09188243448734283,
      "num_tokens": 114629.0,
      "step": 50
    },
    {
      "entropy": 7.324716234207154,
      "epoch": 0.005436931593515224,
      "grad_norm": 0.875,
      "learning_rate": 2.7e-05,
      "loss": 7.2396,
      "mean_token_accuracy": 0.09266308695077896,
      "num_tokens": 125012.0,
      "step": 55
    },
    {
      "entropy": 7.393645143508911,
      "epoch": 0.005931198102016607,
      "grad_norm": 0.87890625,
      "learning_rate": 2.95e-05,
      "loss": 7.1731,
      "mean_token_accuracy": 0.09935653060674668,
      "num_tokens": 135834.0,
      "step": 60
    },
    {
      "entropy": 7.348312473297119,
      "epoch": 0.006425464610517992,
      "grad_norm": 0.89453125,
      "learning_rate": 3.2e-05,
      "loss": 7.0755,
      "mean_token_accuracy": 0.10009140819311142,
      "num_tokens": 147413.0,
      "step": 65
    },
    {
      "entropy": 7.368169450759888,
      "epoch": 0.006919731119019375,
      "grad_norm": 0.9375,
      "learning_rate": 3.4500000000000005e-05,
      "loss": 7.2672,
      "mean_token_accuracy": 0.09652369320392609,
      "num_tokens": 158046.0,
      "step": 70
    },
    {
      "entropy": 7.3670432567596436,
      "epoch": 0.00741399762752076,
      "grad_norm": 0.99609375,
      "learning_rate": 3.7e-05,
      "loss": 7.1792,
      "mean_token_accuracy": 0.09774387776851653,
      "num_tokens": 168564.0,
      "step": 75
    },
    {
      "entropy": 7.313370609283448,
      "epoch": 0.007908264136022143,
      "grad_norm": 0.8515625,
      "learning_rate": 3.95e-05,
      "loss": 7.1885,
      "mean_token_accuracy": 0.09465310201048852,
      "num_tokens": 181168.0,
      "step": 80
    },
    {
      "entropy": 7.35333480834961,
      "epoch": 0.008402530644523527,
      "grad_norm": 0.9296875,
      "learning_rate": 4.2000000000000004e-05,
      "loss": 7.1552,
      "mean_token_accuracy": 0.09859942272305489,
      "num_tokens": 192666.0,
      "step": 85
    },
    {
      "entropy": 7.388514804840088,
      "epoch": 0.008896797153024912,
      "grad_norm": 0.81640625,
      "learning_rate": 4.45e-05,
      "loss": 7.2204,
      "mean_token_accuracy": 0.09482975378632545,
      "num_tokens": 203688.0,
      "step": 90
    },
    {
      "entropy": 7.303425407409668,
      "epoch": 0.009391063661526295,
      "grad_norm": 0.9140625,
      "learning_rate": 4.7000000000000004e-05,
      "loss": 7.0729,
      "mean_token_accuracy": 0.10156495124101639,
      "num_tokens": 214399.0,
      "step": 95
    },
    {
      "entropy": 7.27442512512207,
      "epoch": 0.009885330170027679,
      "grad_norm": 0.765625,
      "learning_rate": 4.9500000000000004e-05,
      "loss": 7.1681,
      "mean_token_accuracy": 0.09274221584200859,
      "num_tokens": 227549.0,
      "step": 100
    },
    {
      "entropy": 7.317437267303466,
      "epoch": 0.010379596678529062,
      "grad_norm": 0.765625,
      "learning_rate": 5.2e-05,
      "loss": 7.0679,
      "mean_token_accuracy": 0.11123422309756278,
      "num_tokens": 239612.0,
      "step": 105
    },
    {
      "entropy": 7.363053894042968,
      "epoch": 0.010873863187030448,
      "grad_norm": 0.96484375,
      "learning_rate": 5.45e-05,
      "loss": 7.1978,
      "mean_token_accuracy": 0.09392422363162041,
      "num_tokens": 251163.0,
      "step": 110
    },
    {
      "entropy": 7.261190605163574,
      "epoch": 0.011368129695531831,
      "grad_norm": 0.87890625,
      "learning_rate": 5.7e-05,
      "loss": 7.0645,
      "mean_token_accuracy": 0.10441159456968307,
      "num_tokens": 263995.0,
      "step": 115
    },
    {
      "entropy": 7.40733699798584,
      "epoch": 0.011862396204033215,
      "grad_norm": 1.0390625,
      "learning_rate": 5.9499999999999996e-05,
      "loss": 7.2354,
      "mean_token_accuracy": 0.08496984019875527,
      "num_tokens": 275103.0,
      "step": 120
    },
    {
      "entropy": 7.273692512512207,
      "epoch": 0.012356662712534598,
      "grad_norm": 0.8515625,
      "learning_rate": 6.2e-05,
      "loss": 7.1369,
      "mean_token_accuracy": 0.0968512661755085,
      "num_tokens": 286838.0,
      "step": 125
    },
    {
      "entropy": 7.423242712020874,
      "epoch": 0.012850929221035983,
      "grad_norm": 1.046875,
      "learning_rate": 6.450000000000001e-05,
      "loss": 7.1665,
      "mean_token_accuracy": 0.09364857673645019,
      "num_tokens": 298750.0,
      "step": 130
    },
    {
      "entropy": 7.250139808654785,
      "epoch": 0.013345195729537367,
      "grad_norm": 0.953125,
      "learning_rate": 6.7e-05,
      "loss": 7.0329,
      "mean_token_accuracy": 0.10625170543789864,
      "num_tokens": 310451.0,
      "step": 135
    },
    {
      "entropy": 7.348578405380249,
      "epoch": 0.01383946223803875,
      "grad_norm": 0.95703125,
      "learning_rate": 6.950000000000001e-05,
      "loss": 7.1935,
      "mean_token_accuracy": 0.09695170894265175,
      "num_tokens": 321232.0,
      "step": 140
    },
    {
      "entropy": 7.488092279434204,
      "epoch": 0.014333728746540134,
      "grad_norm": 0.83984375,
      "learning_rate": 7.2e-05,
      "loss": 7.106,
      "mean_token_accuracy": 0.09741687774658203,
      "num_tokens": 332531.0,
      "step": 145
    },
    {
      "entropy": 7.19592809677124,
      "epoch": 0.01482799525504152,
      "grad_norm": 0.94140625,
      "learning_rate": 7.45e-05,
      "loss": 7.1008,
      "mean_token_accuracy": 0.09623056724667549,
      "num_tokens": 346002.0,
      "step": 150
    },
    {
      "entropy": 7.401388263702392,
      "epoch": 0.015322261763542903,
      "grad_norm": 1.03125,
      "learning_rate": 7.7e-05,
      "loss": 7.0874,
      "mean_token_accuracy": 0.09739502891898155,
      "num_tokens": 356691.0,
      "step": 155
    },
    {
      "entropy": 7.243306970596313,
      "epoch": 0.015816528272044286,
      "grad_norm": 0.94140625,
      "learning_rate": 7.950000000000001e-05,
      "loss": 7.0938,
      "mean_token_accuracy": 0.1009882666170597,
      "num_tokens": 368749.0,
      "step": 160
    },
    {
      "entropy": 7.3369566917419435,
      "epoch": 0.01631079478054567,
      "grad_norm": 1.03125,
      "learning_rate": 8.2e-05,
      "loss": 7.0241,
      "mean_token_accuracy": 0.10501249730587006,
      "num_tokens": 379484.0,
      "step": 165
    },
    {
      "entropy": 7.281102800369263,
      "epoch": 0.016805061289047053,
      "grad_norm": 0.953125,
      "learning_rate": 8.450000000000001e-05,
      "loss": 7.0703,
      "mean_token_accuracy": 0.09354055374860763,
      "num_tokens": 389751.0,
      "step": 170
    },
    {
      "entropy": 7.325433778762817,
      "epoch": 0.017299327797548437,
      "grad_norm": 0.99609375,
      "learning_rate": 8.7e-05,
      "loss": 7.0892,
      "mean_token_accuracy": 0.09684126302599907,
      "num_tokens": 400771.0,
      "step": 175
    },
    {
      "entropy": 7.323175096511841,
      "epoch": 0.017793594306049824,
      "grad_norm": 0.90234375,
      "learning_rate": 8.95e-05,
      "loss": 7.1061,
      "mean_token_accuracy": 0.09781083017587662,
      "num_tokens": 413575.0,
      "step": 180
    },
    {
      "entropy": 7.328048992156982,
      "epoch": 0.018287860814551207,
      "grad_norm": 0.890625,
      "learning_rate": 9.2e-05,
      "loss": 7.0467,
      "mean_token_accuracy": 0.09532680064439773,
      "num_tokens": 424825.0,
      "step": 185
    },
    {
      "entropy": 7.333641099929809,
      "epoch": 0.01878212732305259,
      "grad_norm": 0.93359375,
      "learning_rate": 9.45e-05,
      "loss": 7.1064,
      "mean_token_accuracy": 0.10553916320204734,
      "num_tokens": 435610.0,
      "step": 190
    },
    {
      "entropy": 7.328917694091797,
      "epoch": 0.019276393831553974,
      "grad_norm": 1.0625,
      "learning_rate": 9.7e-05,
      "loss": 7.1187,
      "mean_token_accuracy": 0.0999750517308712,
      "num_tokens": 445855.0,
      "step": 195
    },
    {
      "entropy": 7.2166337966918945,
      "epoch": 0.019770660340055358,
      "grad_norm": 0.98046875,
      "learning_rate": 9.95e-05,
      "loss": 7.0063,
      "mean_token_accuracy": 0.10225539430975913,
      "num_tokens": 456895.0,
      "step": 200
    },
    {
      "entropy": 7.476543855667114,
      "epoch": 0.02026492684855674,
      "grad_norm": 1.234375,
      "learning_rate": 0.000102,
      "loss": 7.1414,
      "mean_token_accuracy": 0.09506701603531838,
      "num_tokens": 467446.0,
      "step": 205
    },
    {
      "entropy": 7.272785186767578,
      "epoch": 0.020759193357058125,
      "grad_norm": 0.921875,
      "learning_rate": 0.00010449999999999999,
      "loss": 7.0607,
      "mean_token_accuracy": 0.09756421521306038,
      "num_tokens": 478988.0,
      "step": 210
    },
    {
      "entropy": 7.290947103500367,
      "epoch": 0.02125345986555951,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000107,
      "loss": 7.0557,
      "mean_token_accuracy": 0.10137153342366219,
      "num_tokens": 491053.0,
      "step": 215
    },
    {
      "entropy": 7.357240819931031,
      "epoch": 0.021747726374060895,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0001095,
      "loss": 7.1334,
      "mean_token_accuracy": 0.09304505437612534,
      "num_tokens": 502279.0,
      "step": 220
    },
    {
      "entropy": 7.198823118209839,
      "epoch": 0.02224199288256228,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000112,
      "loss": 7.0177,
      "mean_token_accuracy": 0.10008044987916946,
      "num_tokens": 513470.0,
      "step": 225
    },
    {
      "entropy": 7.373158121109009,
      "epoch": 0.022736259391063662,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0001145,
      "loss": 7.0871,
      "mean_token_accuracy": 0.096877521276474,
      "num_tokens": 524380.0,
      "step": 230
    },
    {
      "entropy": 7.324968671798706,
      "epoch": 0.023230525899565046,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00011700000000000001,
      "loss": 7.0885,
      "mean_token_accuracy": 0.10163207277655602,
      "num_tokens": 535798.0,
      "step": 235
    },
    {
      "entropy": 7.273069143295288,
      "epoch": 0.02372479240806643,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00011949999999999999,
      "loss": 7.0313,
      "mean_token_accuracy": 0.10062324851751328,
      "num_tokens": 545720.0,
      "step": 240
    },
    {
      "entropy": 7.329104614257813,
      "epoch": 0.024219058916567813,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000122,
      "loss": 7.0595,
      "mean_token_accuracy": 0.09972514137625695,
      "num_tokens": 557110.0,
      "step": 245
    },
    {
      "entropy": 7.1850278854370115,
      "epoch": 0.024713325425069196,
      "grad_norm": 1.109375,
      "learning_rate": 0.0001245,
      "loss": 7.016,
      "mean_token_accuracy": 0.10394479855895042,
      "num_tokens": 568966.0,
      "step": 250
    },
    {
      "entropy": 7.360903692245484,
      "epoch": 0.02520759193357058,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000127,
      "loss": 7.0469,
      "mean_token_accuracy": 0.10300094187259674,
      "num_tokens": 580441.0,
      "step": 255
    },
    {
      "entropy": 7.301402139663696,
      "epoch": 0.025701858442071967,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0001295,
      "loss": 7.0452,
      "mean_token_accuracy": 0.09699605703353882,
      "num_tokens": 592698.0,
      "step": 260
    },
    {
      "entropy": 7.190503692626953,
      "epoch": 0.02619612495057335,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000132,
      "loss": 7.0198,
      "mean_token_accuracy": 0.10279446840286255,
      "num_tokens": 604787.0,
      "step": 265
    },
    {
      "entropy": 7.302949237823486,
      "epoch": 0.026690391459074734,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00013450000000000002,
      "loss": 7.013,
      "mean_token_accuracy": 0.09963318407535553,
      "num_tokens": 616445.0,
      "step": 270
    },
    {
      "entropy": 7.2669110774993895,
      "epoch": 0.027184657967576117,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00013700000000000002,
      "loss": 7.0808,
      "mean_token_accuracy": 0.1002419799566269,
      "num_tokens": 627723.0,
      "step": 275
    },
    {
      "entropy": 7.345779991149902,
      "epoch": 0.0276789244760775,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0001395,
      "loss": 7.1067,
      "mean_token_accuracy": 0.10031414031982422,
      "num_tokens": 638987.0,
      "step": 280
    },
    {
      "entropy": 7.279949617385864,
      "epoch": 0.028173190984578884,
      "grad_norm": 1.296875,
      "learning_rate": 0.00014199999999999998,
      "loss": 6.9384,
      "mean_token_accuracy": 0.10166889876127243,
      "num_tokens": 650626.0,
      "step": 285
    },
    {
      "entropy": 7.149828386306763,
      "epoch": 0.028667457493080268,
      "grad_norm": 1.109375,
      "learning_rate": 0.0001445,
      "loss": 6.9604,
      "mean_token_accuracy": 0.10975496619939804,
      "num_tokens": 661911.0,
      "step": 290
    },
    {
      "entropy": 7.256064605712891,
      "epoch": 0.02916172400158165,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000147,
      "loss": 7.0137,
      "mean_token_accuracy": 0.10477377250790595,
      "num_tokens": 674298.0,
      "step": 295
    },
    {
      "entropy": 7.236632680892944,
      "epoch": 0.02965599051008304,
      "grad_norm": 0.96875,
      "learning_rate": 0.0001495,
      "loss": 6.9431,
      "mean_token_accuracy": 0.10994272455573081,
      "num_tokens": 686000.0,
      "step": 300
    },
    {
      "entropy": 7.216502523422241,
      "epoch": 0.030150257018584422,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000152,
      "loss": 6.9926,
      "mean_token_accuracy": 0.10297218561172486,
      "num_tokens": 698464.0,
      "step": 305
    },
    {
      "entropy": 7.215946102142334,
      "epoch": 0.030644523527085805,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00015450000000000001,
      "loss": 6.9883,
      "mean_token_accuracy": 0.09871925041079521,
      "num_tokens": 709702.0,
      "step": 310
    },
    {
      "entropy": 7.40031623840332,
      "epoch": 0.03113879003558719,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000157,
      "loss": 7.1009,
      "mean_token_accuracy": 0.09918646588921547,
      "num_tokens": 722234.0,
      "step": 315
    },
    {
      "entropy": 7.222815370559692,
      "epoch": 0.03163305654408857,
      "grad_norm": 1.09375,
      "learning_rate": 0.0001595,
      "loss": 7.0443,
      "mean_token_accuracy": 0.10143259838223458,
      "num_tokens": 733001.0,
      "step": 320
    },
    {
      "entropy": 7.251048564910889,
      "epoch": 0.03212732305258996,
      "grad_norm": 1.0625,
      "learning_rate": 0.000162,
      "loss": 7.0047,
      "mean_token_accuracy": 0.09862519353628159,
      "num_tokens": 743315.0,
      "step": 325
    },
    {
      "entropy": 7.253071928024292,
      "epoch": 0.03262158956109134,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00016450000000000001,
      "loss": 6.9549,
      "mean_token_accuracy": 0.09725747108459473,
      "num_tokens": 754386.0,
      "step": 330
    },
    {
      "entropy": 7.24435601234436,
      "epoch": 0.033115856069592726,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00016700000000000002,
      "loss": 7.0694,
      "mean_token_accuracy": 0.10413578823208809,
      "num_tokens": 766504.0,
      "step": 335
    },
    {
      "entropy": 7.090495014190674,
      "epoch": 0.033610122578094107,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00016950000000000003,
      "loss": 6.9017,
      "mean_token_accuracy": 0.10729243233799934,
      "num_tokens": 777039.0,
      "step": 340
    },
    {
      "entropy": 7.284985685348511,
      "epoch": 0.034104389086595494,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00017199999999999998,
      "loss": 7.0025,
      "mean_token_accuracy": 0.10153071135282517,
      "num_tokens": 787890.0,
      "step": 345
    },
    {
      "entropy": 7.234701490402221,
      "epoch": 0.034598655595096874,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00017449999999999999,
      "loss": 6.9532,
      "mean_token_accuracy": 0.10262288972735405,
      "num_tokens": 800129.0,
      "step": 350
    },
    {
      "entropy": 7.139390373229981,
      "epoch": 0.03509292210359826,
      "grad_norm": 0.94140625,
      "learning_rate": 0.000177,
      "loss": 7.0596,
      "mean_token_accuracy": 0.0944197230041027,
      "num_tokens": 812517.0,
      "step": 355
    },
    {
      "entropy": 7.259590196609497,
      "epoch": 0.03558718861209965,
      "grad_norm": 1.0,
      "learning_rate": 0.0001795,
      "loss": 6.9787,
      "mean_token_accuracy": 0.10446251705288886,
      "num_tokens": 824570.0,
      "step": 360
    },
    {
      "entropy": 7.1810768127441404,
      "epoch": 0.03608145512060103,
      "grad_norm": 0.94140625,
      "learning_rate": 0.000182,
      "loss": 6.9823,
      "mean_token_accuracy": 0.1032750204205513,
      "num_tokens": 836506.0,
      "step": 365
    },
    {
      "entropy": 7.251035213470459,
      "epoch": 0.036575721629102415,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0001845,
      "loss": 6.9702,
      "mean_token_accuracy": 0.10457680597901345,
      "num_tokens": 846841.0,
      "step": 370
    },
    {
      "entropy": 7.27037239074707,
      "epoch": 0.037069988137603795,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000187,
      "loss": 7.0054,
      "mean_token_accuracy": 0.106174336373806,
      "num_tokens": 858770.0,
      "step": 375
    },
    {
      "entropy": 7.1212574481964115,
      "epoch": 0.03756425464610518,
      "grad_norm": 1.125,
      "learning_rate": 0.0001895,
      "loss": 6.8786,
      "mean_token_accuracy": 0.10385922640562058,
      "num_tokens": 870184.0,
      "step": 380
    },
    {
      "entropy": 7.172485733032227,
      "epoch": 0.03805852115460656,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000192,
      "loss": 6.9197,
      "mean_token_accuracy": 0.10581747069954872,
      "num_tokens": 881843.0,
      "step": 385
    },
    {
      "entropy": 7.2427778244018555,
      "epoch": 0.03855278766310795,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0001945,
      "loss": 6.9489,
      "mean_token_accuracy": 0.10129198804497719,
      "num_tokens": 893453.0,
      "step": 390
    },
    {
      "entropy": 7.205210494995117,
      "epoch": 0.03904705417160933,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00019700000000000002,
      "loss": 7.0049,
      "mean_token_accuracy": 0.09680559858679771,
      "num_tokens": 904417.0,
      "step": 395
    },
    {
      "entropy": 7.202549695968628,
      "epoch": 0.039541320680110716,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00019950000000000002,
      "loss": 6.9462,
      "mean_token_accuracy": 0.0999035358428955,
      "num_tokens": 915134.0,
      "step": 400
    },
    {
      "entropy": 7.133806133270264,
      "epoch": 0.0400355871886121,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000202,
      "loss": 6.9589,
      "mean_token_accuracy": 0.1034577377140522,
      "num_tokens": 926361.0,
      "step": 405
    },
    {
      "entropy": 7.144542980194092,
      "epoch": 0.04052985369711348,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00020449999999999998,
      "loss": 6.8841,
      "mean_token_accuracy": 0.10176018252968788,
      "num_tokens": 937474.0,
      "step": 410
    },
    {
      "entropy": 7.3199704647064205,
      "epoch": 0.04102412020561487,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000207,
      "loss": 6.947,
      "mean_token_accuracy": 0.10589171275496483,
      "num_tokens": 948897.0,
      "step": 415
    },
    {
      "entropy": 7.157266998291016,
      "epoch": 0.04151838671411625,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002095,
      "loss": 6.9643,
      "mean_token_accuracy": 0.1008329525589943,
      "num_tokens": 960669.0,
      "step": 420
    },
    {
      "entropy": 7.024852180480957,
      "epoch": 0.04201265322261764,
      "grad_norm": 1.0,
      "learning_rate": 0.000212,
      "loss": 6.8152,
      "mean_token_accuracy": 0.10260056778788566,
      "num_tokens": 973001.0,
      "step": 425
    },
    {
      "entropy": 7.119008874893188,
      "epoch": 0.04250691973111902,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002145,
      "loss": 6.8859,
      "mean_token_accuracy": 0.10822405442595481,
      "num_tokens": 986214.0,
      "step": 430
    },
    {
      "entropy": 7.105495595932007,
      "epoch": 0.043001186239620404,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00021700000000000002,
      "loss": 6.8937,
      "mean_token_accuracy": 0.10061700344085693,
      "num_tokens": 997363.0,
      "step": 435
    },
    {
      "entropy": 7.120448970794678,
      "epoch": 0.04349545274812179,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002195,
      "loss": 6.8093,
      "mean_token_accuracy": 0.11288737356662751,
      "num_tokens": 1009543.0,
      "step": 440
    },
    {
      "entropy": 7.0541246891021725,
      "epoch": 0.04398971925662317,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000222,
      "loss": 6.9215,
      "mean_token_accuracy": 0.10269275158643723,
      "num_tokens": 1020561.0,
      "step": 445
    },
    {
      "entropy": 7.185726308822632,
      "epoch": 0.04448398576512456,
      "grad_norm": 0.984375,
      "learning_rate": 0.0002245,
      "loss": 6.8725,
      "mean_token_accuracy": 0.10224331989884376,
      "num_tokens": 1032587.0,
      "step": 450
    },
    {
      "entropy": 7.159319496154785,
      "epoch": 0.04497825227362594,
      "grad_norm": 1.140625,
      "learning_rate": 0.00022700000000000002,
      "loss": 6.8784,
      "mean_token_accuracy": 0.1019284375011921,
      "num_tokens": 1044249.0,
      "step": 455
    },
    {
      "entropy": 7.076584959030152,
      "epoch": 0.045472518782127325,
      "grad_norm": 1.328125,
      "learning_rate": 0.00022950000000000002,
      "loss": 6.9215,
      "mean_token_accuracy": 0.10558762773871422,
      "num_tokens": 1055415.0,
      "step": 460
    },
    {
      "entropy": 7.1408079147338865,
      "epoch": 0.045966785290628705,
      "grad_norm": 1.03125,
      "learning_rate": 0.00023200000000000003,
      "loss": 6.8942,
      "mean_token_accuracy": 0.10558660700917244,
      "num_tokens": 1066593.0,
      "step": 465
    },
    {
      "entropy": 7.15621657371521,
      "epoch": 0.04646105179913009,
      "grad_norm": 1.109375,
      "learning_rate": 0.00023449999999999998,
      "loss": 6.8924,
      "mean_token_accuracy": 0.10691260844469071,
      "num_tokens": 1076966.0,
      "step": 470
    },
    {
      "entropy": 7.067881631851196,
      "epoch": 0.04695531830763147,
      "grad_norm": 1.03125,
      "learning_rate": 0.000237,
      "loss": 6.9491,
      "mean_token_accuracy": 0.09959092512726783,
      "num_tokens": 1089927.0,
      "step": 475
    },
    {
      "entropy": 7.091880464553833,
      "epoch": 0.04744958481613286,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002395,
      "loss": 6.8869,
      "mean_token_accuracy": 0.10327862426638604,
      "num_tokens": 1101920.0,
      "step": 480
    },
    {
      "entropy": 7.182226657867432,
      "epoch": 0.047943851324634246,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000242,
      "loss": 6.9376,
      "mean_token_accuracy": 0.10818986669182777,
      "num_tokens": 1115078.0,
      "step": 485
    },
    {
      "entropy": 7.180230808258057,
      "epoch": 0.048438117833135626,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002445,
      "loss": 7.0315,
      "mean_token_accuracy": 0.0969414807856083,
      "num_tokens": 1127138.0,
      "step": 490
    },
    {
      "entropy": 7.088583707809448,
      "epoch": 0.04893238434163701,
      "grad_norm": 1.1875,
      "learning_rate": 0.000247,
      "loss": 6.815,
      "mean_token_accuracy": 0.10144753381609917,
      "num_tokens": 1138390.0,
      "step": 495
    },
    {
      "entropy": 7.003582715988159,
      "epoch": 0.04942665085013839,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002495,
      "loss": 6.78,
      "mean_token_accuracy": 0.11048796996474267,
      "num_tokens": 1149124.0,
      "step": 500
    },
    {
      "entropy": 7.101625585556031,
      "epoch": 0.04992091735863978,
      "grad_norm": 1.109375,
      "learning_rate": 0.000252,
      "loss": 6.8585,
      "mean_token_accuracy": 0.10951191484928131,
      "num_tokens": 1161091.0,
      "step": 505
    },
    {
      "entropy": 7.118922424316406,
      "epoch": 0.05041518386714116,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0002545,
      "loss": 6.8355,
      "mean_token_accuracy": 0.10461836084723472,
      "num_tokens": 1171919.0,
      "step": 510
    },
    {
      "entropy": 7.051087188720703,
      "epoch": 0.05090945037564255,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000257,
      "loss": 6.8633,
      "mean_token_accuracy": 0.10401171222329139,
      "num_tokens": 1183688.0,
      "step": 515
    },
    {
      "entropy": 7.0168451309204105,
      "epoch": 0.051403716884143934,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002595,
      "loss": 6.7819,
      "mean_token_accuracy": 0.10449134409427643,
      "num_tokens": 1195323.0,
      "step": 520
    },
    {
      "entropy": 7.09809103012085,
      "epoch": 0.051897983392645314,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000262,
      "loss": 6.8999,
      "mean_token_accuracy": 0.10218222662806511,
      "num_tokens": 1206782.0,
      "step": 525
    },
    {
      "entropy": 6.965949535369873,
      "epoch": 0.0523922499011467,
      "grad_norm": 1.0625,
      "learning_rate": 0.00026450000000000003,
      "loss": 6.8013,
      "mean_token_accuracy": 0.10973932594060898,
      "num_tokens": 1218102.0,
      "step": 530
    },
    {
      "entropy": 7.017365598678589,
      "epoch": 0.05288651640964808,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00026700000000000004,
      "loss": 6.7659,
      "mean_token_accuracy": 0.10771052315831184,
      "num_tokens": 1230199.0,
      "step": 535
    },
    {
      "entropy": 7.0721368312835695,
      "epoch": 0.05338078291814947,
      "grad_norm": 1.03125,
      "learning_rate": 0.00026950000000000005,
      "loss": 6.8896,
      "mean_token_accuracy": 0.10399755015969277,
      "num_tokens": 1242127.0,
      "step": 540
    },
    {
      "entropy": 7.027034282684326,
      "epoch": 0.05387504942665085,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00027200000000000005,
      "loss": 6.7655,
      "mean_token_accuracy": 0.10667505711317063,
      "num_tokens": 1252329.0,
      "step": 545
    },
    {
      "entropy": 6.865018033981324,
      "epoch": 0.054369315935152235,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002745,
      "loss": 6.7666,
      "mean_token_accuracy": 0.11077538877725601,
      "num_tokens": 1264832.0,
      "step": 550
    },
    {
      "entropy": 7.053894948959351,
      "epoch": 0.054863582443653615,
      "grad_norm": 1.0625,
      "learning_rate": 0.000277,
      "loss": 6.8414,
      "mean_token_accuracy": 0.10259129181504249,
      "num_tokens": 1276708.0,
      "step": 555
    },
    {
      "entropy": 7.017140913009643,
      "epoch": 0.055357848952155,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002795,
      "loss": 6.7446,
      "mean_token_accuracy": 0.11817902401089668,
      "num_tokens": 1288089.0,
      "step": 560
    },
    {
      "entropy": 7.102619647979736,
      "epoch": 0.05585211546065639,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00028199999999999997,
      "loss": 6.9011,
      "mean_token_accuracy": 0.10063079670071602,
      "num_tokens": 1299411.0,
      "step": 565
    },
    {
      "entropy": 7.038193321228027,
      "epoch": 0.05634638196915777,
      "grad_norm": 1.015625,
      "learning_rate": 0.0002845,
      "loss": 6.8678,
      "mean_token_accuracy": 0.10497712716460228,
      "num_tokens": 1311368.0,
      "step": 570
    },
    {
      "entropy": 6.996581506729126,
      "epoch": 0.056840648477659156,
      "grad_norm": 1.140625,
      "learning_rate": 0.000287,
      "loss": 6.7798,
      "mean_token_accuracy": 0.1104745551943779,
      "num_tokens": 1321962.0,
      "step": 575
    },
    {
      "entropy": 6.984814691543579,
      "epoch": 0.057334914986160536,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002895,
      "loss": 6.7845,
      "mean_token_accuracy": 0.10723584592342376,
      "num_tokens": 1335150.0,
      "step": 580
    },
    {
      "entropy": 7.0060882568359375,
      "epoch": 0.05782918149466192,
      "grad_norm": 1.015625,
      "learning_rate": 0.000292,
      "loss": 6.8106,
      "mean_token_accuracy": 0.10954170823097228,
      "num_tokens": 1346986.0,
      "step": 585
    },
    {
      "entropy": 7.04237003326416,
      "epoch": 0.0583234480031633,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002945,
      "loss": 6.7456,
      "mean_token_accuracy": 0.11029022559523582,
      "num_tokens": 1357820.0,
      "step": 590
    },
    {
      "entropy": 6.898483562469482,
      "epoch": 0.05881771451166469,
      "grad_norm": 1.203125,
      "learning_rate": 0.000297,
      "loss": 6.7381,
      "mean_token_accuracy": 0.1155846431851387,
      "num_tokens": 1368269.0,
      "step": 595
    },
    {
      "entropy": 6.914667129516602,
      "epoch": 0.05931198102016608,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002995,
      "loss": 6.7699,
      "mean_token_accuracy": 0.10781971290707588,
      "num_tokens": 1380364.0,
      "step": 600
    },
    {
      "entropy": 6.9423737049102785,
      "epoch": 0.05980624752866746,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000302,
      "loss": 6.7359,
      "mean_token_accuracy": 0.11458505317568779,
      "num_tokens": 1390667.0,
      "step": 605
    },
    {
      "entropy": 6.890728282928467,
      "epoch": 0.060300514037168844,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003045,
      "loss": 6.6885,
      "mean_token_accuracy": 0.10985699743032455,
      "num_tokens": 1401452.0,
      "step": 610
    },
    {
      "entropy": 6.968847942352295,
      "epoch": 0.060794780545670224,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000307,
      "loss": 6.6936,
      "mean_token_accuracy": 0.11100370958447456,
      "num_tokens": 1412947.0,
      "step": 615
    },
    {
      "entropy": 6.867507553100586,
      "epoch": 0.06128904705417161,
      "grad_norm": 0.953125,
      "learning_rate": 0.0003095,
      "loss": 6.7476,
      "mean_token_accuracy": 0.10398394390940666,
      "num_tokens": 1425123.0,
      "step": 620
    },
    {
      "entropy": 6.928701877593994,
      "epoch": 0.06178331356267299,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000312,
      "loss": 6.7218,
      "mean_token_accuracy": 0.10982921719551086,
      "num_tokens": 1436538.0,
      "step": 625
    },
    {
      "entropy": 6.933251285552979,
      "epoch": 0.06227758007117438,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003145,
      "loss": 6.77,
      "mean_token_accuracy": 0.10577303692698478,
      "num_tokens": 1448269.0,
      "step": 630
    },
    {
      "entropy": 6.900527620315552,
      "epoch": 0.06277184657967576,
      "grad_norm": 1.1875,
      "learning_rate": 0.000317,
      "loss": 6.7432,
      "mean_token_accuracy": 0.11349172964692116,
      "num_tokens": 1459729.0,
      "step": 635
    },
    {
      "entropy": 6.960564184188843,
      "epoch": 0.06326611308817714,
      "grad_norm": 1.125,
      "learning_rate": 0.0003195,
      "loss": 6.7581,
      "mean_token_accuracy": 0.10724205747246743,
      "num_tokens": 1470684.0,
      "step": 640
    },
    {
      "entropy": 6.889400863647461,
      "epoch": 0.06376037959667853,
      "grad_norm": 0.9375,
      "learning_rate": 0.000322,
      "loss": 6.7232,
      "mean_token_accuracy": 0.10671663433313369,
      "num_tokens": 1483571.0,
      "step": 645
    },
    {
      "entropy": 6.8848834991455075,
      "epoch": 0.06425464610517992,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00032450000000000003,
      "loss": 6.6891,
      "mean_token_accuracy": 0.11692152470350266,
      "num_tokens": 1494671.0,
      "step": 650
    },
    {
      "entropy": 6.905150508880615,
      "epoch": 0.06474891261368129,
      "grad_norm": 1.296875,
      "learning_rate": 0.00032700000000000003,
      "loss": 6.7095,
      "mean_token_accuracy": 0.1126905307173729,
      "num_tokens": 1505435.0,
      "step": 655
    },
    {
      "entropy": 6.948776197433472,
      "epoch": 0.06524317912218268,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00032950000000000004,
      "loss": 6.803,
      "mean_token_accuracy": 0.10531418174505233,
      "num_tokens": 1518510.0,
      "step": 660
    },
    {
      "entropy": 6.9253791809082035,
      "epoch": 0.06573744563068407,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00033200000000000005,
      "loss": 6.7145,
      "mean_token_accuracy": 0.11463218927383423,
      "num_tokens": 1530218.0,
      "step": 665
    },
    {
      "entropy": 6.91641206741333,
      "epoch": 0.06623171213918545,
      "grad_norm": 1.078125,
      "learning_rate": 0.00033450000000000005,
      "loss": 6.7582,
      "mean_token_accuracy": 0.11466718763113022,
      "num_tokens": 1542486.0,
      "step": 670
    },
    {
      "entropy": 6.8444969177246096,
      "epoch": 0.06672597864768683,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000337,
      "loss": 6.7533,
      "mean_token_accuracy": 0.10930409133434296,
      "num_tokens": 1553873.0,
      "step": 675
    },
    {
      "entropy": 6.942787122726441,
      "epoch": 0.06722024515618821,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003395,
      "loss": 6.7047,
      "mean_token_accuracy": 0.1153480589389801,
      "num_tokens": 1565343.0,
      "step": 680
    },
    {
      "entropy": 6.903558826446533,
      "epoch": 0.0677145116646896,
      "grad_norm": 1.078125,
      "learning_rate": 0.000342,
      "loss": 6.7608,
      "mean_token_accuracy": 0.10652155056595802,
      "num_tokens": 1577061.0,
      "step": 685
    },
    {
      "entropy": 6.872377204895019,
      "epoch": 0.06820877817319099,
      "grad_norm": 1.0625,
      "learning_rate": 0.00034449999999999997,
      "loss": 6.7529,
      "mean_token_accuracy": 0.11244799494743347,
      "num_tokens": 1588083.0,
      "step": 690
    },
    {
      "entropy": 6.879619789123535,
      "epoch": 0.06870304468169237,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000347,
      "loss": 6.7159,
      "mean_token_accuracy": 0.10822473615407943,
      "num_tokens": 1599796.0,
      "step": 695
    },
    {
      "entropy": 6.90326623916626,
      "epoch": 0.06919731119019375,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003495,
      "loss": 6.7681,
      "mean_token_accuracy": 0.10771963074803352,
      "num_tokens": 1611994.0,
      "step": 700
    },
    {
      "entropy": 6.872850322723389,
      "epoch": 0.06969157769869513,
      "grad_norm": 1.0625,
      "learning_rate": 0.000352,
      "loss": 6.7022,
      "mean_token_accuracy": 0.10573551878333091,
      "num_tokens": 1623280.0,
      "step": 705
    },
    {
      "entropy": 6.8690471172332765,
      "epoch": 0.07018584420719652,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003545,
      "loss": 6.7681,
      "mean_token_accuracy": 0.10785022675991059,
      "num_tokens": 1634989.0,
      "step": 710
    },
    {
      "entropy": 6.947969198226929,
      "epoch": 0.07068011071569791,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000357,
      "loss": 6.7925,
      "mean_token_accuracy": 0.10478616431355477,
      "num_tokens": 1647042.0,
      "step": 715
    },
    {
      "entropy": 6.809376001358032,
      "epoch": 0.0711743772241993,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003595,
      "loss": 6.6855,
      "mean_token_accuracy": 0.1154635414481163,
      "num_tokens": 1659245.0,
      "step": 720
    },
    {
      "entropy": 6.81630311012268,
      "epoch": 0.07166864373270067,
      "grad_norm": 1.140625,
      "learning_rate": 0.000362,
      "loss": 6.6819,
      "mean_token_accuracy": 0.10983811244368553,
      "num_tokens": 1670271.0,
      "step": 725
    },
    {
      "entropy": 6.873712110519409,
      "epoch": 0.07216291024120206,
      "grad_norm": 0.96875,
      "learning_rate": 0.0003645,
      "loss": 6.735,
      "mean_token_accuracy": 0.11080555841326714,
      "num_tokens": 1682648.0,
      "step": 730
    },
    {
      "entropy": 6.822260570526123,
      "epoch": 0.07265717674970344,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000367,
      "loss": 6.7573,
      "mean_token_accuracy": 0.1052017480134964,
      "num_tokens": 1694542.0,
      "step": 735
    },
    {
      "entropy": 6.849833726882935,
      "epoch": 0.07315144325820483,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003695,
      "loss": 6.6493,
      "mean_token_accuracy": 0.10704348310828209,
      "num_tokens": 1705737.0,
      "step": 740
    },
    {
      "entropy": 6.862112379074096,
      "epoch": 0.0736457097667062,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000372,
      "loss": 6.71,
      "mean_token_accuracy": 0.10965840369462967,
      "num_tokens": 1718474.0,
      "step": 745
    },
    {
      "entropy": 6.826199579238891,
      "epoch": 0.07413997627520759,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003745,
      "loss": 6.6873,
      "mean_token_accuracy": 0.1082968644797802,
      "num_tokens": 1729360.0,
      "step": 750
    },
    {
      "entropy": 6.8160795211792,
      "epoch": 0.07463424278370898,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000377,
      "loss": 6.6964,
      "mean_token_accuracy": 0.1107470378279686,
      "num_tokens": 1739654.0,
      "step": 755
    },
    {
      "entropy": 6.836647272109985,
      "epoch": 0.07512850929221036,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003795,
      "loss": 6.6939,
      "mean_token_accuracy": 0.11239867880940438,
      "num_tokens": 1752158.0,
      "step": 760
    },
    {
      "entropy": 6.834174489974975,
      "epoch": 0.07562277580071175,
      "grad_norm": 1.171875,
      "learning_rate": 0.000382,
      "loss": 6.6923,
      "mean_token_accuracy": 0.11612112373113632,
      "num_tokens": 1764545.0,
      "step": 765
    },
    {
      "entropy": 6.821158313751221,
      "epoch": 0.07611704230921312,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003845,
      "loss": 6.6814,
      "mean_token_accuracy": 0.1128039114177227,
      "num_tokens": 1775422.0,
      "step": 770
    },
    {
      "entropy": 6.840568494796753,
      "epoch": 0.07661130881771451,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00038700000000000003,
      "loss": 6.7088,
      "mean_token_accuracy": 0.10114984959363937,
      "num_tokens": 1787833.0,
      "step": 775
    },
    {
      "entropy": 6.74714674949646,
      "epoch": 0.0771055753262159,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00038950000000000003,
      "loss": 6.5723,
      "mean_token_accuracy": 0.1219009131193161,
      "num_tokens": 1798508.0,
      "step": 780
    },
    {
      "entropy": 6.753849983215332,
      "epoch": 0.07759984183471728,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039200000000000004,
      "loss": 6.6209,
      "mean_token_accuracy": 0.11469837054610252,
      "num_tokens": 1809873.0,
      "step": 785
    },
    {
      "entropy": 6.821059226989746,
      "epoch": 0.07809410834321866,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00039450000000000005,
      "loss": 6.7014,
      "mean_token_accuracy": 0.11237634271383286,
      "num_tokens": 1820876.0,
      "step": 790
    },
    {
      "entropy": 6.7968470573425295,
      "epoch": 0.07858837485172004,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00039700000000000005,
      "loss": 6.663,
      "mean_token_accuracy": 0.10904834121465683,
      "num_tokens": 1833233.0,
      "step": 795
    },
    {
      "entropy": 6.812273120880127,
      "epoch": 0.07908264136022143,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003995,
      "loss": 6.7481,
      "mean_token_accuracy": 0.10785818174481392,
      "num_tokens": 1844520.0,
      "step": 800
    },
    {
      "entropy": 6.765308666229248,
      "epoch": 0.07957690786872282,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000402,
      "loss": 6.6292,
      "mean_token_accuracy": 0.1142186164855957,
      "num_tokens": 1856814.0,
      "step": 805
    },
    {
      "entropy": 6.732009649276733,
      "epoch": 0.0800711743772242,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004045,
      "loss": 6.6196,
      "mean_token_accuracy": 0.11811698004603385,
      "num_tokens": 1867581.0,
      "step": 810
    },
    {
      "entropy": 6.84562578201294,
      "epoch": 0.08056544088572558,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00040699999999999997,
      "loss": 6.6907,
      "mean_token_accuracy": 0.10590304806828499,
      "num_tokens": 1879302.0,
      "step": 815
    },
    {
      "entropy": 6.756962633132934,
      "epoch": 0.08105970739422697,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004095,
      "loss": 6.6571,
      "mean_token_accuracy": 0.10640354081988335,
      "num_tokens": 1891697.0,
      "step": 820
    },
    {
      "entropy": 6.759096002578735,
      "epoch": 0.08155397390272835,
      "grad_norm": 1.078125,
      "learning_rate": 0.000412,
      "loss": 6.5979,
      "mean_token_accuracy": 0.11474953293800354,
      "num_tokens": 1903544.0,
      "step": 825
    },
    {
      "entropy": 6.774603033065796,
      "epoch": 0.08204824041122974,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004145,
      "loss": 6.6545,
      "mean_token_accuracy": 0.11101827323436737,
      "num_tokens": 1915286.0,
      "step": 830
    },
    {
      "entropy": 6.688508129119873,
      "epoch": 0.08254250691973111,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000417,
      "loss": 6.5346,
      "mean_token_accuracy": 0.11861704438924789,
      "num_tokens": 1926361.0,
      "step": 835
    },
    {
      "entropy": 6.751745128631592,
      "epoch": 0.0830367734282325,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004195,
      "loss": 6.6299,
      "mean_token_accuracy": 0.11083293855190277,
      "num_tokens": 1938540.0,
      "step": 840
    },
    {
      "entropy": 6.763704109191894,
      "epoch": 0.08353103993673389,
      "grad_norm": 0.9140625,
      "learning_rate": 0.000422,
      "loss": 6.6603,
      "mean_token_accuracy": 0.10724062249064445,
      "num_tokens": 1950532.0,
      "step": 845
    },
    {
      "entropy": 6.8038843154907225,
      "epoch": 0.08402530644523527,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004245,
      "loss": 6.6778,
      "mean_token_accuracy": 0.10919453650712967,
      "num_tokens": 1960899.0,
      "step": 850
    },
    {
      "entropy": 6.68118052482605,
      "epoch": 0.08451957295373666,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000427,
      "loss": 6.6496,
      "mean_token_accuracy": 0.11730073392391205,
      "num_tokens": 1972278.0,
      "step": 855
    },
    {
      "entropy": 6.725075197219849,
      "epoch": 0.08501383946223803,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004295,
      "loss": 6.6692,
      "mean_token_accuracy": 0.1092053584754467,
      "num_tokens": 1985079.0,
      "step": 860
    },
    {
      "entropy": 6.635829877853394,
      "epoch": 0.08550810597073942,
      "grad_norm": 1.15625,
      "learning_rate": 0.000432,
      "loss": 6.7065,
      "mean_token_accuracy": 0.10706321522593498,
      "num_tokens": 1996467.0,
      "step": 865
    },
    {
      "entropy": 6.8284436702728275,
      "epoch": 0.08600237247924081,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004345,
      "loss": 6.5832,
      "mean_token_accuracy": 0.11370615139603615,
      "num_tokens": 2007339.0,
      "step": 870
    },
    {
      "entropy": 6.677596378326416,
      "epoch": 0.0864966389877422,
      "grad_norm": 1.015625,
      "learning_rate": 0.000437,
      "loss": 6.5879,
      "mean_token_accuracy": 0.1196245089173317,
      "num_tokens": 2019962.0,
      "step": 875
    },
    {
      "entropy": 6.747453260421753,
      "epoch": 0.08699090549624358,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004395,
      "loss": 6.5866,
      "mean_token_accuracy": 0.11758355721831322,
      "num_tokens": 2031251.0,
      "step": 880
    },
    {
      "entropy": 6.661469602584839,
      "epoch": 0.08748517200474495,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000442,
      "loss": 6.6013,
      "mean_token_accuracy": 0.11161932870745658,
      "num_tokens": 2042412.0,
      "step": 885
    },
    {
      "entropy": 6.6397772312164305,
      "epoch": 0.08797943851324634,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004445,
      "loss": 6.6221,
      "mean_token_accuracy": 0.11210491955280304,
      "num_tokens": 2054423.0,
      "step": 890
    },
    {
      "entropy": 6.686139965057373,
      "epoch": 0.08847370502174773,
      "grad_norm": 1.046875,
      "learning_rate": 0.000447,
      "loss": 6.595,
      "mean_token_accuracy": 0.10692231729626656,
      "num_tokens": 2065657.0,
      "step": 895
    },
    {
      "entropy": 6.717379331588745,
      "epoch": 0.08896797153024912,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044950000000000003,
      "loss": 6.5827,
      "mean_token_accuracy": 0.11325386464595795,
      "num_tokens": 2076305.0,
      "step": 900
    },
    {
      "entropy": 6.697859430313111,
      "epoch": 0.08946223803875049,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045200000000000004,
      "loss": 6.5783,
      "mean_token_accuracy": 0.11316154897212982,
      "num_tokens": 2087734.0,
      "step": 905
    },
    {
      "entropy": 6.595232200622559,
      "epoch": 0.08995650454725188,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045450000000000004,
      "loss": 6.4953,
      "mean_token_accuracy": 0.12035880610346794,
      "num_tokens": 2098016.0,
      "step": 910
    },
    {
      "entropy": 6.665337467193604,
      "epoch": 0.09045077105575326,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00045700000000000005,
      "loss": 6.6678,
      "mean_token_accuracy": 0.10886707603931427,
      "num_tokens": 2110293.0,
      "step": 915
    },
    {
      "entropy": 6.673768377304077,
      "epoch": 0.09094503756425465,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045950000000000006,
      "loss": 6.4485,
      "mean_token_accuracy": 0.11682527139782906,
      "num_tokens": 2120441.0,
      "step": 920
    },
    {
      "entropy": 6.539246511459351,
      "epoch": 0.09143930407275604,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000462,
      "loss": 6.5683,
      "mean_token_accuracy": 0.11887063086032867,
      "num_tokens": 2131520.0,
      "step": 925
    },
    {
      "entropy": 6.664091300964356,
      "epoch": 0.09193357058125741,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004645,
      "loss": 6.5796,
      "mean_token_accuracy": 0.11637971252202987,
      "num_tokens": 2145005.0,
      "step": 930
    },
    {
      "entropy": 6.605041790008545,
      "epoch": 0.0924278370897588,
      "grad_norm": 1.0625,
      "learning_rate": 0.000467,
      "loss": 6.5338,
      "mean_token_accuracy": 0.1211809828877449,
      "num_tokens": 2156608.0,
      "step": 935
    },
    {
      "entropy": 6.601588439941406,
      "epoch": 0.09292210359826018,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004695,
      "loss": 6.5318,
      "mean_token_accuracy": 0.11969754993915557,
      "num_tokens": 2168341.0,
      "step": 940
    },
    {
      "entropy": 6.625442409515381,
      "epoch": 0.09341637010676157,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000472,
      "loss": 6.5258,
      "mean_token_accuracy": 0.11236160099506379,
      "num_tokens": 2178704.0,
      "step": 945
    },
    {
      "entropy": 6.704224395751953,
      "epoch": 0.09391063661526294,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004745,
      "loss": 6.6906,
      "mean_token_accuracy": 0.10928287953138352,
      "num_tokens": 2190884.0,
      "step": 950
    },
    {
      "entropy": 6.682482194900513,
      "epoch": 0.09440490312376433,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000477,
      "loss": 6.598,
      "mean_token_accuracy": 0.1167626291513443,
      "num_tokens": 2201822.0,
      "step": 955
    },
    {
      "entropy": 6.629432439804077,
      "epoch": 0.09489916963226572,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004795,
      "loss": 6.5926,
      "mean_token_accuracy": 0.11254436299204826,
      "num_tokens": 2213362.0,
      "step": 960
    },
    {
      "entropy": 6.692664337158203,
      "epoch": 0.0953934361407671,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000482,
      "loss": 6.602,
      "mean_token_accuracy": 0.11311232969164849,
      "num_tokens": 2224366.0,
      "step": 965
    },
    {
      "entropy": 6.46690821647644,
      "epoch": 0.09588770264926849,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004845,
      "loss": 6.4535,
      "mean_token_accuracy": 0.1190331868827343,
      "num_tokens": 2234645.0,
      "step": 970
    },
    {
      "entropy": 6.678108978271484,
      "epoch": 0.09638196915776986,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000487,
      "loss": 6.5498,
      "mean_token_accuracy": 0.12018159404397011,
      "num_tokens": 2245904.0,
      "step": 975
    },
    {
      "entropy": 6.4710277080535885,
      "epoch": 0.09687623566627125,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004895,
      "loss": 6.49,
      "mean_token_accuracy": 0.11942838132381439,
      "num_tokens": 2257753.0,
      "step": 980
    },
    {
      "entropy": 6.551316976547241,
      "epoch": 0.09737050217477264,
      "grad_norm": 1.03125,
      "learning_rate": 0.000492,
      "loss": 6.4953,
      "mean_token_accuracy": 0.1244705080986023,
      "num_tokens": 2269009.0,
      "step": 985
    },
    {
      "entropy": 6.693281459808349,
      "epoch": 0.09786476868327403,
      "grad_norm": 1.0,
      "learning_rate": 0.0004945,
      "loss": 6.5493,
      "mean_token_accuracy": 0.11529032737016678,
      "num_tokens": 2279840.0,
      "step": 990
    },
    {
      "entropy": 6.533772134780884,
      "epoch": 0.0983590351917754,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000497,
      "loss": 6.5154,
      "mean_token_accuracy": 0.11877169013023377,
      "num_tokens": 2290577.0,
      "step": 995
    },
    {
      "entropy": 6.553924179077148,
      "epoch": 0.09885330170027679,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004995,
      "loss": 6.5274,
      "mean_token_accuracy": 0.12048136815428734,
      "num_tokens": 2301341.0,
      "step": 1000
    },
    {
      "entropy": 6.566124486923218,
      "epoch": 0.09934756820877817,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999999982287888,
      "loss": 6.5896,
      "mean_token_accuracy": 0.11639286801218987,
      "num_tokens": 2312505.0,
      "step": 1005
    },
    {
      "entropy": 6.718512153625488,
      "epoch": 0.09984183471727956,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999999910332434,
      "loss": 6.5236,
      "mean_token_accuracy": 0.11419004276394844,
      "num_tokens": 2323443.0,
      "step": 1010
    },
    {
      "entropy": 6.4791271686553955,
      "epoch": 0.10033610122578095,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999999783026635,
      "loss": 6.5439,
      "mean_token_accuracy": 0.11673406213521957,
      "num_tokens": 2335686.0,
      "step": 1015
    },
    {
      "entropy": 6.514018297195435,
      "epoch": 0.10083036773428232,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999999600370491,
      "loss": 6.5178,
      "mean_token_accuracy": 0.11814445331692695,
      "num_tokens": 2347273.0,
      "step": 1020
    },
    {
      "entropy": 6.552177906036377,
      "epoch": 0.1013246342427837,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999999362364008,
      "loss": 6.4427,
      "mean_token_accuracy": 0.12624893859028816,
      "num_tokens": 2358011.0,
      "step": 1025
    },
    {
      "entropy": 6.5004723072052,
      "epoch": 0.1018189007512851,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999999069007191,
      "loss": 6.5815,
      "mean_token_accuracy": 0.11536967605352402,
      "num_tokens": 2371390.0,
      "step": 1030
    },
    {
      "entropy": 6.596110677719116,
      "epoch": 0.10231316725978648,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999998720300048,
      "loss": 6.5271,
      "mean_token_accuracy": 0.12005563005805016,
      "num_tokens": 2381887.0,
      "step": 1035
    },
    {
      "entropy": 6.551437664031982,
      "epoch": 0.10280743376828787,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004999998316242589,
      "loss": 6.4663,
      "mean_token_accuracy": 0.11999196782708169,
      "num_tokens": 2393632.0,
      "step": 1040
    },
    {
      "entropy": 6.530908870697021,
      "epoch": 0.10330170027678924,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999997856834821,
      "loss": 6.4778,
      "mean_token_accuracy": 0.12028345614671707,
      "num_tokens": 2405686.0,
      "step": 1045
    },
    {
      "entropy": 6.457987642288208,
      "epoch": 0.10379596678529063,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004999997342076757,
      "loss": 6.455,
      "mean_token_accuracy": 0.11674800738692284,
      "num_tokens": 2417459.0,
      "step": 1050
    },
    {
      "entropy": 6.636799621582031,
      "epoch": 0.10429023329379201,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004999996771968408,
      "loss": 6.4995,
      "mean_token_accuracy": 0.11853041127324104,
      "num_tokens": 2428689.0,
      "step": 1055
    },
    {
      "entropy": 6.509030675888061,
      "epoch": 0.1047844998022934,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999996146509791,
      "loss": 6.5295,
      "mean_token_accuracy": 0.11135807186365128,
      "num_tokens": 2439437.0,
      "step": 1060
    },
    {
      "entropy": 6.580507850646972,
      "epoch": 0.10527876631079477,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999995465700919,
      "loss": 6.5104,
      "mean_token_accuracy": 0.11756298467516899,
      "num_tokens": 2450576.0,
      "step": 1065
    },
    {
      "entropy": 6.587932443618774,
      "epoch": 0.10577303281929616,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999994729541809,
      "loss": 6.6509,
      "mean_token_accuracy": 0.11502083018422127,
      "num_tokens": 2462684.0,
      "step": 1070
    },
    {
      "entropy": 6.633237361907959,
      "epoch": 0.10626729932779755,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004999993938032481,
      "loss": 6.5244,
      "mean_token_accuracy": 0.12519583106040955,
      "num_tokens": 2474193.0,
      "step": 1075
    },
    {
      "entropy": 6.394119358062744,
      "epoch": 0.10676156583629894,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999993091172952,
      "loss": 6.4655,
      "mean_token_accuracy": 0.11539505198597907,
      "num_tokens": 2486264.0,
      "step": 1080
    },
    {
      "entropy": 6.555998420715332,
      "epoch": 0.10725583234480032,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004999992188963245,
      "loss": 6.4541,
      "mean_token_accuracy": 0.12652762085199357,
      "num_tokens": 2497668.0,
      "step": 1085
    },
    {
      "entropy": 6.488944244384766,
      "epoch": 0.1077500988533017,
      "grad_norm": 0.96875,
      "learning_rate": 0.000499999123140338,
      "loss": 6.5113,
      "mean_token_accuracy": 0.11666428670287132,
      "num_tokens": 2509635.0,
      "step": 1090
    },
    {
      "entropy": 6.545294523239136,
      "epoch": 0.10824436536180308,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999990218493382,
      "loss": 6.5051,
      "mean_token_accuracy": 0.12337316423654557,
      "num_tokens": 2521980.0,
      "step": 1095
    },
    {
      "entropy": 6.497365427017212,
      "epoch": 0.10873863187030447,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004999989150233276,
      "loss": 6.4639,
      "mean_token_accuracy": 0.11867155879735947,
      "num_tokens": 2533750.0,
      "step": 1100
    },
    {
      "entropy": 6.565949201583862,
      "epoch": 0.10923289837880586,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999988026623088,
      "loss": 6.5618,
      "mean_token_accuracy": 0.11584960371255874,
      "num_tokens": 2545620.0,
      "step": 1105
    },
    {
      "entropy": 6.454707479476928,
      "epoch": 0.10972716488730723,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999986847662843,
      "loss": 6.3302,
      "mean_token_accuracy": 0.13472548872232437,
      "num_tokens": 2556463.0,
      "step": 1110
    },
    {
      "entropy": 6.503189325332642,
      "epoch": 0.11022143139580862,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999985613352576,
      "loss": 6.5358,
      "mean_token_accuracy": 0.12201443389058113,
      "num_tokens": 2567396.0,
      "step": 1115
    },
    {
      "entropy": 6.544538831710815,
      "epoch": 0.11071569790431,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004999984323692311,
      "loss": 6.48,
      "mean_token_accuracy": 0.123195381462574,
      "num_tokens": 2579208.0,
      "step": 1120
    },
    {
      "entropy": 6.412357044219971,
      "epoch": 0.11120996441281139,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999982978682085,
      "loss": 6.4248,
      "mean_token_accuracy": 0.12385009601712227,
      "num_tokens": 2590940.0,
      "step": 1125
    },
    {
      "entropy": 6.558205413818359,
      "epoch": 0.11170423092131278,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004999981578321927,
      "loss": 6.51,
      "mean_token_accuracy": 0.1274452067911625,
      "num_tokens": 2602491.0,
      "step": 1130
    },
    {
      "entropy": 6.439068698883057,
      "epoch": 0.11219849742981415,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004999980122611874,
      "loss": 6.3438,
      "mean_token_accuracy": 0.12809207662940025,
      "num_tokens": 2612139.0,
      "step": 1135
    },
    {
      "entropy": 6.445886325836182,
      "epoch": 0.11269276393831554,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000499997861155196,
      "loss": 6.5104,
      "mean_token_accuracy": 0.12374393939971924,
      "num_tokens": 2624906.0,
      "step": 1140
    },
    {
      "entropy": 6.474299335479737,
      "epoch": 0.11318703044681692,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004999977045142223,
      "loss": 6.4204,
      "mean_token_accuracy": 0.12185963913798332,
      "num_tokens": 2636046.0,
      "step": 1145
    },
    {
      "entropy": 6.521690130233765,
      "epoch": 0.11368129695531831,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999975423382703,
      "loss": 6.4288,
      "mean_token_accuracy": 0.13111222088336943,
      "num_tokens": 2645617.0,
      "step": 1150
    },
    {
      "entropy": 6.534403610229492,
      "epoch": 0.11417556346381968,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004999973746273438,
      "loss": 6.4529,
      "mean_token_accuracy": 0.12487476840615272,
      "num_tokens": 2656780.0,
      "step": 1155
    },
    {
      "entropy": 6.454224348068237,
      "epoch": 0.11466982997232107,
      "grad_norm": 1.03125,
      "learning_rate": 0.000499997201381447,
      "loss": 6.438,
      "mean_token_accuracy": 0.1251967266201973,
      "num_tokens": 2668078.0,
      "step": 1160
    },
    {
      "entropy": 6.491998386383057,
      "epoch": 0.11516409648082246,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999970226005841,
      "loss": 6.5242,
      "mean_token_accuracy": 0.11767346188426017,
      "num_tokens": 2679049.0,
      "step": 1165
    },
    {
      "entropy": 6.53035717010498,
      "epoch": 0.11565836298932385,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004999968382847595,
      "loss": 6.4624,
      "mean_token_accuracy": 0.12308069095015525,
      "num_tokens": 2690165.0,
      "step": 1170
    },
    {
      "entropy": 6.536982250213623,
      "epoch": 0.11615262949782523,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999966484339778,
      "loss": 6.5002,
      "mean_token_accuracy": 0.116629458963871,
      "num_tokens": 2701965.0,
      "step": 1175
    },
    {
      "entropy": 6.402723407745361,
      "epoch": 0.1166468960063266,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999964530482437,
      "loss": 6.421,
      "mean_token_accuracy": 0.12439692988991738,
      "num_tokens": 2713479.0,
      "step": 1180
    },
    {
      "entropy": 6.465710687637329,
      "epoch": 0.11714116251482799,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999962521275619,
      "loss": 6.436,
      "mean_token_accuracy": 0.12075797766447068,
      "num_tokens": 2725629.0,
      "step": 1185
    },
    {
      "entropy": 6.268183088302612,
      "epoch": 0.11763542902332938,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999960456719374,
      "loss": 6.3492,
      "mean_token_accuracy": 0.12263342142105102,
      "num_tokens": 2736620.0,
      "step": 1190
    },
    {
      "entropy": 6.464852428436279,
      "epoch": 0.11812969553183077,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999958336813753,
      "loss": 6.3325,
      "mean_token_accuracy": 0.1303306147456169,
      "num_tokens": 2747118.0,
      "step": 1195
    },
    {
      "entropy": 6.372828674316406,
      "epoch": 0.11862396204033215,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999956161558807,
      "loss": 6.3799,
      "mean_token_accuracy": 0.1274876520037651,
      "num_tokens": 2758371.0,
      "step": 1200
    },
    {
      "entropy": 6.391465854644776,
      "epoch": 0.11911822854883353,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004999953930954591,
      "loss": 6.3852,
      "mean_token_accuracy": 0.12252886295318603,
      "num_tokens": 2770575.0,
      "step": 1205
    },
    {
      "entropy": 6.448058319091797,
      "epoch": 0.11961249505733491,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004999951645001159,
      "loss": 6.3455,
      "mean_token_accuracy": 0.1301012769341469,
      "num_tokens": 2783103.0,
      "step": 1210
    },
    {
      "entropy": 6.436516189575196,
      "epoch": 0.1201067615658363,
      "grad_norm": 0.80078125,
      "learning_rate": 0.0004999949303698568,
      "loss": 6.4665,
      "mean_token_accuracy": 0.12414569854736328,
      "num_tokens": 2796636.0,
      "step": 1215
    },
    {
      "entropy": 6.406609773635864,
      "epoch": 0.12060102807433769,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004999946907046875,
      "loss": 6.322,
      "mean_token_accuracy": 0.13122917115688323,
      "num_tokens": 2807798.0,
      "step": 1220
    },
    {
      "entropy": 6.362497234344483,
      "epoch": 0.12109529458283906,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004999944455046139,
      "loss": 6.439,
      "mean_token_accuracy": 0.12562692314386367,
      "num_tokens": 2819255.0,
      "step": 1225
    },
    {
      "entropy": 6.454278612136841,
      "epoch": 0.12158956109134045,
      "grad_norm": 1.03125,
      "learning_rate": 0.000499994194769642,
      "loss": 6.4012,
      "mean_token_accuracy": 0.1260200709104538,
      "num_tokens": 2829952.0,
      "step": 1230
    },
    {
      "entropy": 6.518359088897705,
      "epoch": 0.12208382759984183,
      "grad_norm": 1.046875,
      "learning_rate": 0.000499993938499778,
      "loss": 6.4773,
      "mean_token_accuracy": 0.1282096892595291,
      "num_tokens": 2841056.0,
      "step": 1235
    },
    {
      "entropy": 6.377371597290039,
      "epoch": 0.12257809410834322,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999936766950283,
      "loss": 6.3282,
      "mean_token_accuracy": 0.13203255608677864,
      "num_tokens": 2852045.0,
      "step": 1240
    },
    {
      "entropy": 6.4568657875061035,
      "epoch": 0.12307236061684461,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999934093553992,
      "loss": 6.434,
      "mean_token_accuracy": 0.12442487627267837,
      "num_tokens": 2863440.0,
      "step": 1245
    },
    {
      "entropy": 6.430741882324218,
      "epoch": 0.12356662712534598,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999931364808972,
      "loss": 6.4369,
      "mean_token_accuracy": 0.1217712052166462,
      "num_tokens": 2874017.0,
      "step": 1250
    },
    {
      "entropy": 6.364642333984375,
      "epoch": 0.12406089363384737,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999928580715293,
      "loss": 6.3657,
      "mean_token_accuracy": 0.1264870472252369,
      "num_tokens": 2885677.0,
      "step": 1255
    },
    {
      "entropy": 6.433004665374756,
      "epoch": 0.12455516014234876,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000499992574127302,
      "loss": 6.4812,
      "mean_token_accuracy": 0.1225816085934639,
      "num_tokens": 2897825.0,
      "step": 1260
    },
    {
      "entropy": 6.524859094619751,
      "epoch": 0.12504942665085014,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004999922846482226,
      "loss": 6.5257,
      "mean_token_accuracy": 0.1220851257443428,
      "num_tokens": 2909304.0,
      "step": 1265
    },
    {
      "entropy": 6.429646253585815,
      "epoch": 0.12554369315935152,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004999919896342981,
      "loss": 6.4588,
      "mean_token_accuracy": 0.12074072510004044,
      "num_tokens": 2920918.0,
      "step": 1270
    },
    {
      "entropy": 6.501003408432007,
      "epoch": 0.12603795966785292,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004999916890855358,
      "loss": 6.4073,
      "mean_token_accuracy": 0.12426696866750717,
      "num_tokens": 2932294.0,
      "step": 1275
    },
    {
      "entropy": 6.363391876220703,
      "epoch": 0.1265322261763543,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999913830019431,
      "loss": 6.3237,
      "mean_token_accuracy": 0.13384533002972604,
      "num_tokens": 2943372.0,
      "step": 1280
    },
    {
      "entropy": 6.311191463470459,
      "epoch": 0.12702649268485566,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999910713835272,
      "loss": 6.3047,
      "mean_token_accuracy": 0.13015900626778604,
      "num_tokens": 2953807.0,
      "step": 1285
    },
    {
      "entropy": 6.342788410186768,
      "epoch": 0.12752075919335706,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999907542302962,
      "loss": 6.2438,
      "mean_token_accuracy": 0.13418981581926345,
      "num_tokens": 2964372.0,
      "step": 1290
    },
    {
      "entropy": 6.376524639129639,
      "epoch": 0.12801502570185844,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999904315422577,
      "loss": 6.3365,
      "mean_token_accuracy": 0.1274750992655754,
      "num_tokens": 2975537.0,
      "step": 1295
    },
    {
      "entropy": 6.296396970748901,
      "epoch": 0.12850929221035984,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999901033194198,
      "loss": 6.3172,
      "mean_token_accuracy": 0.12614937424659728,
      "num_tokens": 2986094.0,
      "step": 1300
    },
    {
      "entropy": 6.412167549133301,
      "epoch": 0.1290035587188612,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999897695617904,
      "loss": 6.4284,
      "mean_token_accuracy": 0.12951384037733077,
      "num_tokens": 2996023.0,
      "step": 1305
    },
    {
      "entropy": 6.477496385574341,
      "epoch": 0.12949782522736258,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004999894302693778,
      "loss": 6.4201,
      "mean_token_accuracy": 0.12168471738696099,
      "num_tokens": 3007584.0,
      "step": 1310
    },
    {
      "entropy": 6.470839118957519,
      "epoch": 0.12999209173586398,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999890854421902,
      "loss": 6.4722,
      "mean_token_accuracy": 0.11812516897916794,
      "num_tokens": 3018995.0,
      "step": 1315
    },
    {
      "entropy": 6.263270092010498,
      "epoch": 0.13048635824436536,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004999887350802362,
      "loss": 6.2577,
      "mean_token_accuracy": 0.13282620459794997,
      "num_tokens": 3030576.0,
      "step": 1320
    },
    {
      "entropy": 6.371448612213134,
      "epoch": 0.13098062475286676,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999883791835246,
      "loss": 6.3224,
      "mean_token_accuracy": 0.12991052865982056,
      "num_tokens": 3040765.0,
      "step": 1325
    },
    {
      "entropy": 6.39703049659729,
      "epoch": 0.13147489126136813,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004999880177520637,
      "loss": 6.3713,
      "mean_token_accuracy": 0.12290074527263642,
      "num_tokens": 3053912.0,
      "step": 1330
    },
    {
      "entropy": 6.269509172439575,
      "epoch": 0.1319691577698695,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004999876507858628,
      "loss": 6.358,
      "mean_token_accuracy": 0.12943168208003045,
      "num_tokens": 3065921.0,
      "step": 1335
    },
    {
      "entropy": 6.491255092620849,
      "epoch": 0.1324634242783709,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004999872782849308,
      "loss": 6.4118,
      "mean_token_accuracy": 0.1217479057610035,
      "num_tokens": 3078643.0,
      "step": 1340
    },
    {
      "entropy": 6.297278165817261,
      "epoch": 0.13295769078687228,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999869002492768,
      "loss": 6.3262,
      "mean_token_accuracy": 0.12477339208126068,
      "num_tokens": 3090300.0,
      "step": 1345
    },
    {
      "entropy": 6.313536787033081,
      "epoch": 0.13345195729537365,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004999865166789102,
      "loss": 6.2574,
      "mean_token_accuracy": 0.13334870636463164,
      "num_tokens": 3102307.0,
      "step": 1350
    },
    {
      "entropy": 6.430685234069824,
      "epoch": 0.13394622380387505,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999861275738404,
      "loss": 6.4038,
      "mean_token_accuracy": 0.12200917750597,
      "num_tokens": 3113817.0,
      "step": 1355
    },
    {
      "entropy": 6.360499048233033,
      "epoch": 0.13444049031237643,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999857329340769,
      "loss": 6.2516,
      "mean_token_accuracy": 0.12839845269918443,
      "num_tokens": 3124719.0,
      "step": 1360
    },
    {
      "entropy": 6.246741867065429,
      "epoch": 0.13493475682087783,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004999853327596295,
      "loss": 6.2873,
      "mean_token_accuracy": 0.1282256707549095,
      "num_tokens": 3136136.0,
      "step": 1365
    },
    {
      "entropy": 6.370300054550171,
      "epoch": 0.1354290233293792,
      "grad_norm": 0.90234375,
      "learning_rate": 0.000499984927050508,
      "loss": 6.3012,
      "mean_token_accuracy": 0.1290183626115322,
      "num_tokens": 3148977.0,
      "step": 1370
    },
    {
      "entropy": 6.377018070220947,
      "epoch": 0.13592328983788057,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999845158067223,
      "loss": 6.3579,
      "mean_token_accuracy": 0.1252500332891941,
      "num_tokens": 3160520.0,
      "step": 1375
    },
    {
      "entropy": 6.366725778579712,
      "epoch": 0.13641755634638197,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004999840990282828,
      "loss": 6.3426,
      "mean_token_accuracy": 0.12962477132678032,
      "num_tokens": 3172895.0,
      "step": 1380
    },
    {
      "entropy": 6.324902153015136,
      "epoch": 0.13691182285488335,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999836767151994,
      "loss": 6.3544,
      "mean_token_accuracy": 0.13052070513367653,
      "num_tokens": 3183960.0,
      "step": 1385
    },
    {
      "entropy": 6.330108976364135,
      "epoch": 0.13740608936338475,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999832488674827,
      "loss": 6.3886,
      "mean_token_accuracy": 0.1273997098207474,
      "num_tokens": 3194276.0,
      "step": 1390
    },
    {
      "entropy": 6.366259002685547,
      "epoch": 0.13790035587188612,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999828154851432,
      "loss": 6.2558,
      "mean_token_accuracy": 0.13061744794249536,
      "num_tokens": 3206084.0,
      "step": 1395
    },
    {
      "entropy": 6.301025104522705,
      "epoch": 0.1383946223803875,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999823765681915,
      "loss": 6.2972,
      "mean_token_accuracy": 0.13314774185419082,
      "num_tokens": 3217941.0,
      "step": 1400
    },
    {
      "entropy": 6.338549423217773,
      "epoch": 0.1388888888888889,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004999819321166385,
      "loss": 6.2742,
      "mean_token_accuracy": 0.13190385103225707,
      "num_tokens": 3229088.0,
      "step": 1405
    },
    {
      "entropy": 6.25566258430481,
      "epoch": 0.13938315539739027,
      "grad_norm": 1.0,
      "learning_rate": 0.000499981482130495,
      "loss": 6.2205,
      "mean_token_accuracy": 0.13236010000109671,
      "num_tokens": 3240708.0,
      "step": 1410
    },
    {
      "entropy": 6.294620752334595,
      "epoch": 0.13987742190589167,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004999810266097721,
      "loss": 6.3444,
      "mean_token_accuracy": 0.1264272078871727,
      "num_tokens": 3251040.0,
      "step": 1415
    },
    {
      "entropy": 6.291800880432129,
      "epoch": 0.14037168841439304,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999805655544811,
      "loss": 6.2912,
      "mean_token_accuracy": 0.13168664127588273,
      "num_tokens": 3262655.0,
      "step": 1420
    },
    {
      "entropy": 6.291258907318115,
      "epoch": 0.14086595492289442,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999800989646333,
      "loss": 6.2228,
      "mean_token_accuracy": 0.13803994432091712,
      "num_tokens": 3273160.0,
      "step": 1425
    },
    {
      "entropy": 6.300457239151001,
      "epoch": 0.14136022143139582,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004999796268402401,
      "loss": 6.2783,
      "mean_token_accuracy": 0.1296892285346985,
      "num_tokens": 3285932.0,
      "step": 1430
    },
    {
      "entropy": 6.249773216247559,
      "epoch": 0.1418544879398972,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004999791491813132,
      "loss": 6.2971,
      "mean_token_accuracy": 0.12768914625048639,
      "num_tokens": 3298834.0,
      "step": 1435
    },
    {
      "entropy": 6.409033393859863,
      "epoch": 0.1423487544483986,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999786659878643,
      "loss": 6.2799,
      "mean_token_accuracy": 0.12664505764842032,
      "num_tokens": 3310473.0,
      "step": 1440
    },
    {
      "entropy": 6.182092523574829,
      "epoch": 0.14284302095689996,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004999781772599055,
      "loss": 6.2497,
      "mean_token_accuracy": 0.13217792585492133,
      "num_tokens": 3322662.0,
      "step": 1445
    },
    {
      "entropy": 6.391679096221924,
      "epoch": 0.14333728746540134,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004999776829974484,
      "loss": 6.3475,
      "mean_token_accuracy": 0.132525734603405,
      "num_tokens": 3334321.0,
      "step": 1450
    },
    {
      "entropy": 6.246479558944702,
      "epoch": 0.14383155397390274,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004999771832005055,
      "loss": 6.2512,
      "mean_token_accuracy": 0.13095758333802224,
      "num_tokens": 3345130.0,
      "step": 1455
    },
    {
      "entropy": 6.2692060470581055,
      "epoch": 0.1443258204824041,
      "grad_norm": 0.984375,
      "learning_rate": 0.000499976677869089,
      "loss": 6.2271,
      "mean_token_accuracy": 0.13088599741458892,
      "num_tokens": 3355601.0,
      "step": 1460
    },
    {
      "entropy": 6.194102716445923,
      "epoch": 0.14482008699090548,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004999761670032115,
      "loss": 6.2295,
      "mean_token_accuracy": 0.13360995799303055,
      "num_tokens": 3367075.0,
      "step": 1465
    },
    {
      "entropy": 6.32319974899292,
      "epoch": 0.14531435349940688,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004999756506028851,
      "loss": 6.1937,
      "mean_token_accuracy": 0.13329043239355087,
      "num_tokens": 3378888.0,
      "step": 1470
    },
    {
      "entropy": 6.248707866668701,
      "epoch": 0.14580862000790826,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004999751286681229,
      "loss": 6.2325,
      "mean_token_accuracy": 0.1328907147049904,
      "num_tokens": 3390705.0,
      "step": 1475
    },
    {
      "entropy": 6.288841915130615,
      "epoch": 0.14630288651640966,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004999746011989376,
      "loss": 6.249,
      "mean_token_accuracy": 0.1373384565114975,
      "num_tokens": 3401710.0,
      "step": 1480
    },
    {
      "entropy": 6.346017694473266,
      "epoch": 0.14679715302491103,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999740681953422,
      "loss": 6.3789,
      "mean_token_accuracy": 0.1264251612126827,
      "num_tokens": 3413467.0,
      "step": 1485
    },
    {
      "entropy": 6.331368350982666,
      "epoch": 0.1472914195334124,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999735296573499,
      "loss": 6.3007,
      "mean_token_accuracy": 0.13292821869254112,
      "num_tokens": 3425885.0,
      "step": 1490
    },
    {
      "entropy": 6.329574871063232,
      "epoch": 0.1477856860419138,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004999729855849738,
      "loss": 6.3151,
      "mean_token_accuracy": 0.12490678578615189,
      "num_tokens": 3437952.0,
      "step": 1495
    },
    {
      "entropy": 6.267677640914917,
      "epoch": 0.14827995255041518,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004999724359782274,
      "loss": 6.1911,
      "mean_token_accuracy": 0.1455492317676544,
      "num_tokens": 3449680.0,
      "step": 1500
    },
    {
      "entropy": 6.317681789398193,
      "epoch": 0.14877421905891658,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999718808371241,
      "loss": 6.3432,
      "mean_token_accuracy": 0.12537492960691451,
      "num_tokens": 3460699.0,
      "step": 1505
    },
    {
      "entropy": 6.338581228256226,
      "epoch": 0.14926848556741795,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004999713201616776,
      "loss": 6.2238,
      "mean_token_accuracy": 0.13232341259717942,
      "num_tokens": 3471919.0,
      "step": 1510
    },
    {
      "entropy": 6.166346454620362,
      "epoch": 0.14976275207591933,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000499970753951902,
      "loss": 6.3229,
      "mean_token_accuracy": 0.12744686827063562,
      "num_tokens": 3484110.0,
      "step": 1515
    },
    {
      "entropy": 6.412110614776611,
      "epoch": 0.15025701858442073,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999701822078107,
      "loss": 6.2352,
      "mean_token_accuracy": 0.12983164712786674,
      "num_tokens": 3496103.0,
      "step": 1520
    },
    {
      "entropy": 6.272561883926391,
      "epoch": 0.1507512850929221,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000499969604929418,
      "loss": 6.3098,
      "mean_token_accuracy": 0.13165246173739434,
      "num_tokens": 3507291.0,
      "step": 1525
    },
    {
      "entropy": 6.302027225494385,
      "epoch": 0.1512455516014235,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004999690221167382,
      "loss": 6.3173,
      "mean_token_accuracy": 0.12720877900719643,
      "num_tokens": 3519564.0,
      "step": 1530
    },
    {
      "entropy": 6.204291439056396,
      "epoch": 0.15173981810992487,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004999684337697855,
      "loss": 6.1724,
      "mean_token_accuracy": 0.13716719225049018,
      "num_tokens": 3530679.0,
      "step": 1535
    },
    {
      "entropy": 6.2982782363891605,
      "epoch": 0.15223408461842625,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004999678398885746,
      "loss": 6.2473,
      "mean_token_accuracy": 0.1318476304411888,
      "num_tokens": 3542662.0,
      "step": 1540
    },
    {
      "entropy": 6.265805768966675,
      "epoch": 0.15272835112692765,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004999672404731197,
      "loss": 6.3499,
      "mean_token_accuracy": 0.12811536714434624,
      "num_tokens": 3554760.0,
      "step": 1545
    },
    {
      "entropy": 6.249004793167114,
      "epoch": 0.15322261763542902,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000499966635523436,
      "loss": 6.1602,
      "mean_token_accuracy": 0.13934924975037574,
      "num_tokens": 3565604.0,
      "step": 1550
    },
    {
      "entropy": 6.277536392211914,
      "epoch": 0.1537168841439304,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999660250395381,
      "loss": 6.3141,
      "mean_token_accuracy": 0.12867220491170883,
      "num_tokens": 3577535.0,
      "step": 1555
    },
    {
      "entropy": 6.295968151092529,
      "epoch": 0.1542111506524318,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000499965409021441,
      "loss": 6.3235,
      "mean_token_accuracy": 0.12822821363806725,
      "num_tokens": 3590458.0,
      "step": 1560
    },
    {
      "entropy": 6.375429725646972,
      "epoch": 0.15470541716093317,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00049996478746916,
      "loss": 6.2851,
      "mean_token_accuracy": 0.13392977118492128,
      "num_tokens": 3600916.0,
      "step": 1565
    },
    {
      "entropy": 6.222699928283691,
      "epoch": 0.15519968366943457,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004999641603827104,
      "loss": 6.3037,
      "mean_token_accuracy": 0.12990911155939103,
      "num_tokens": 3612794.0,
      "step": 1570
    },
    {
      "entropy": 6.27130880355835,
      "epoch": 0.15569395017793594,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999635277621073,
      "loss": 6.2037,
      "mean_token_accuracy": 0.1420078307390213,
      "num_tokens": 3624186.0,
      "step": 1575
    },
    {
      "entropy": 6.199323654174805,
      "epoch": 0.15618821668643731,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004999628896073668,
      "loss": 6.2448,
      "mean_token_accuracy": 0.12877988070249557,
      "num_tokens": 3636041.0,
      "step": 1580
    },
    {
      "entropy": 6.304298067092896,
      "epoch": 0.15668248319493872,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999622459185042,
      "loss": 6.3071,
      "mean_token_accuracy": 0.12442853599786759,
      "num_tokens": 3646859.0,
      "step": 1585
    },
    {
      "entropy": 6.2107603549957275,
      "epoch": 0.1571767497034401,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004999615966955354,
      "loss": 6.119,
      "mean_token_accuracy": 0.14032085835933686,
      "num_tokens": 3657853.0,
      "step": 1590
    },
    {
      "entropy": 6.324536895751953,
      "epoch": 0.1576710162119415,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999609419384764,
      "loss": 6.2253,
      "mean_token_accuracy": 0.134918150305748,
      "num_tokens": 3668341.0,
      "step": 1595
    },
    {
      "entropy": 6.249271965026855,
      "epoch": 0.15816528272044286,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999602816473434,
      "loss": 6.2563,
      "mean_token_accuracy": 0.13581479713320732,
      "num_tokens": 3679020.0,
      "step": 1600
    },
    {
      "entropy": 6.1892296314239506,
      "epoch": 0.15865954922894424,
      "grad_norm": 0.875,
      "learning_rate": 0.0004999596158221526,
      "loss": 6.1613,
      "mean_token_accuracy": 0.13848758935928346,
      "num_tokens": 3691121.0,
      "step": 1605
    },
    {
      "entropy": 6.234239339828491,
      "epoch": 0.15915381573744564,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004999589444629203,
      "loss": 6.2286,
      "mean_token_accuracy": 0.13591036275029184,
      "num_tokens": 3702584.0,
      "step": 1610
    },
    {
      "entropy": 6.20807032585144,
      "epoch": 0.159648082245947,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999582675696631,
      "loss": 6.177,
      "mean_token_accuracy": 0.1353621505200863,
      "num_tokens": 3713085.0,
      "step": 1615
    },
    {
      "entropy": 6.320692300796509,
      "epoch": 0.1601423487544484,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999575851423976,
      "loss": 6.2929,
      "mean_token_accuracy": 0.12485695034265518,
      "num_tokens": 3724318.0,
      "step": 1620
    },
    {
      "entropy": 6.2754554271698,
      "epoch": 0.16063661526294978,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004999568971811407,
      "loss": 6.1999,
      "mean_token_accuracy": 0.14148797765374183,
      "num_tokens": 3734288.0,
      "step": 1625
    },
    {
      "entropy": 6.201929521560669,
      "epoch": 0.16113088177145116,
      "grad_norm": 0.953125,
      "learning_rate": 0.000499956203685909,
      "loss": 6.1941,
      "mean_token_accuracy": 0.13511522933840753,
      "num_tokens": 3744645.0,
      "step": 1630
    },
    {
      "entropy": 6.282017421722412,
      "epoch": 0.16162514827995256,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00049995550465672,
      "loss": 6.1854,
      "mean_token_accuracy": 0.13890446051955224,
      "num_tokens": 3755660.0,
      "step": 1635
    },
    {
      "entropy": 6.125180339813232,
      "epoch": 0.16211941478845393,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999548000935906,
      "loss": 6.1599,
      "mean_token_accuracy": 0.14167107045650482,
      "num_tokens": 3767614.0,
      "step": 1640
    },
    {
      "entropy": 6.254905891418457,
      "epoch": 0.16261368129695533,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999540899965383,
      "loss": 6.308,
      "mean_token_accuracy": 0.1335205040872097,
      "num_tokens": 3780750.0,
      "step": 1645
    },
    {
      "entropy": 6.242093515396118,
      "epoch": 0.1631079478054567,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004999533743655803,
      "loss": 6.2965,
      "mean_token_accuracy": 0.1281206913292408,
      "num_tokens": 3793195.0,
      "step": 1650
    },
    {
      "entropy": 6.24614839553833,
      "epoch": 0.16360221431395808,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999526532007346,
      "loss": 6.2606,
      "mean_token_accuracy": 0.13350182473659516,
      "num_tokens": 3804256.0,
      "step": 1655
    },
    {
      "entropy": 6.289435911178589,
      "epoch": 0.16409648082245948,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999519265020187,
      "loss": 6.2237,
      "mean_token_accuracy": 0.12859297171235085,
      "num_tokens": 3815403.0,
      "step": 1660
    },
    {
      "entropy": 6.08995943069458,
      "epoch": 0.16459074733096085,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999511942694504,
      "loss": 6.128,
      "mean_token_accuracy": 0.14284673556685448,
      "num_tokens": 3826682.0,
      "step": 1665
    },
    {
      "entropy": 6.215432214736938,
      "epoch": 0.16508501383946222,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999504565030479,
      "loss": 6.139,
      "mean_token_accuracy": 0.13716638311743737,
      "num_tokens": 3837780.0,
      "step": 1670
    },
    {
      "entropy": 6.164974164962769,
      "epoch": 0.16557928034796363,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004999497132028293,
      "loss": 6.2072,
      "mean_token_accuracy": 0.1344188041985035,
      "num_tokens": 3848782.0,
      "step": 1675
    },
    {
      "entropy": 6.244879055023193,
      "epoch": 0.166073546856465,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999489643688127,
      "loss": 6.1649,
      "mean_token_accuracy": 0.13939150497317315,
      "num_tokens": 3859220.0,
      "step": 1680
    },
    {
      "entropy": 6.102788019180298,
      "epoch": 0.1665678133649664,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999482100010169,
      "loss": 6.0808,
      "mean_token_accuracy": 0.1425606593489647,
      "num_tokens": 3869794.0,
      "step": 1685
    },
    {
      "entropy": 6.161266660690307,
      "epoch": 0.16706207987346777,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004999474500994601,
      "loss": 6.2184,
      "mean_token_accuracy": 0.1334006331861019,
      "num_tokens": 3881133.0,
      "step": 1690
    },
    {
      "entropy": 6.2152763366699215,
      "epoch": 0.16755634638196915,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999466846641611,
      "loss": 6.1567,
      "mean_token_accuracy": 0.1364052601158619,
      "num_tokens": 3891600.0,
      "step": 1695
    },
    {
      "entropy": 6.224546241760254,
      "epoch": 0.16805061289047055,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004999459136951389,
      "loss": 6.2529,
      "mean_token_accuracy": 0.13441457897424697,
      "num_tokens": 3903287.0,
      "step": 1700
    },
    {
      "entropy": 6.245505905151367,
      "epoch": 0.16854487939897192,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999451371924123,
      "loss": 6.1635,
      "mean_token_accuracy": 0.14323566406965255,
      "num_tokens": 3914497.0,
      "step": 1705
    },
    {
      "entropy": 6.2008990287780765,
      "epoch": 0.16903914590747332,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004999443551560003,
      "loss": 6.1996,
      "mean_token_accuracy": 0.13277414664626122,
      "num_tokens": 3925536.0,
      "step": 1710
    },
    {
      "entropy": 6.179309701919555,
      "epoch": 0.1695334124159747,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004999435675859224,
      "loss": 6.1432,
      "mean_token_accuracy": 0.1426541492342949,
      "num_tokens": 3936705.0,
      "step": 1715
    },
    {
      "entropy": 6.117032384872436,
      "epoch": 0.17002767892447607,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999427744821977,
      "loss": 6.0382,
      "mean_token_accuracy": 0.1389991022646427,
      "num_tokens": 3948422.0,
      "step": 1720
    },
    {
      "entropy": 6.144522857666016,
      "epoch": 0.17052194543297747,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000499941975844846,
      "loss": 6.1085,
      "mean_token_accuracy": 0.14725376516580582,
      "num_tokens": 3958697.0,
      "step": 1725
    },
    {
      "entropy": 6.112959909439087,
      "epoch": 0.17101621194147884,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004999411716738866,
      "loss": 6.1208,
      "mean_token_accuracy": 0.13783455789089202,
      "num_tokens": 3970255.0,
      "step": 1730
    },
    {
      "entropy": 6.206806802749634,
      "epoch": 0.17151047844998024,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999403619693397,
      "loss": 6.2172,
      "mean_token_accuracy": 0.13266652897000314,
      "num_tokens": 3980892.0,
      "step": 1735
    },
    {
      "entropy": 6.101211071014404,
      "epoch": 0.17200474495848161,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999395467312249,
      "loss": 6.1532,
      "mean_token_accuracy": 0.13273056447505951,
      "num_tokens": 3991375.0,
      "step": 1740
    },
    {
      "entropy": 6.304367208480835,
      "epoch": 0.172499011466983,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999387259595623,
      "loss": 6.3042,
      "mean_token_accuracy": 0.12940156012773513,
      "num_tokens": 4003352.0,
      "step": 1745
    },
    {
      "entropy": 6.152772092819214,
      "epoch": 0.1729932779754844,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004999378996543723,
      "loss": 6.1745,
      "mean_token_accuracy": 0.13783681988716126,
      "num_tokens": 4014429.0,
      "step": 1750
    },
    {
      "entropy": 6.22617883682251,
      "epoch": 0.17348754448398576,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999370678156749,
      "loss": 6.1056,
      "mean_token_accuracy": 0.14252740070223807,
      "num_tokens": 4025213.0,
      "step": 1755
    },
    {
      "entropy": 6.1182598114013675,
      "epoch": 0.17398181099248716,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999362304434909,
      "loss": 6.0983,
      "mean_token_accuracy": 0.1411632165312767,
      "num_tokens": 4036244.0,
      "step": 1760
    },
    {
      "entropy": 6.218455600738525,
      "epoch": 0.17447607750098854,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999353875378406,
      "loss": 6.1392,
      "mean_token_accuracy": 0.13833620101213456,
      "num_tokens": 4047944.0,
      "step": 1765
    },
    {
      "entropy": 6.125010585784912,
      "epoch": 0.1749703440094899,
      "grad_norm": 0.921875,
      "learning_rate": 0.000499934539098745,
      "loss": 6.1127,
      "mean_token_accuracy": 0.13938926681876182,
      "num_tokens": 4058865.0,
      "step": 1770
    },
    {
      "entropy": 6.112168598175049,
      "epoch": 0.1754646105179913,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004999336851262247,
      "loss": 6.1605,
      "mean_token_accuracy": 0.13652415424585343,
      "num_tokens": 4070627.0,
      "step": 1775
    },
    {
      "entropy": 6.221715450286865,
      "epoch": 0.17595887702649268,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004999328256203009,
      "loss": 6.2029,
      "mean_token_accuracy": 0.13283938243985177,
      "num_tokens": 4081406.0,
      "step": 1780
    },
    {
      "entropy": 6.202430009841919,
      "epoch": 0.17645314353499406,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004999319605809947,
      "loss": 6.1763,
      "mean_token_accuracy": 0.1280331313610077,
      "num_tokens": 4092957.0,
      "step": 1785
    },
    {
      "entropy": 6.209701681137085,
      "epoch": 0.17694741004349546,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999310900083274,
      "loss": 6.232,
      "mean_token_accuracy": 0.13681847751140594,
      "num_tokens": 4104381.0,
      "step": 1790
    },
    {
      "entropy": 6.180554962158203,
      "epoch": 0.17744167655199683,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999302139023204,
      "loss": 6.1452,
      "mean_token_accuracy": 0.13721615672111512,
      "num_tokens": 4117190.0,
      "step": 1795
    },
    {
      "entropy": 6.11452693939209,
      "epoch": 0.17793594306049823,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999293322629951,
      "loss": 6.0752,
      "mean_token_accuracy": 0.14162272289395333,
      "num_tokens": 4127870.0,
      "step": 1800
    },
    {
      "entropy": 6.098488092422485,
      "epoch": 0.1784302095689996,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999284450903736,
      "loss": 6.1272,
      "mean_token_accuracy": 0.14127160534262656,
      "num_tokens": 4139240.0,
      "step": 1805
    },
    {
      "entropy": 6.223893737792968,
      "epoch": 0.17892447607750098,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999275523844772,
      "loss": 6.0978,
      "mean_token_accuracy": 0.14117365032434465,
      "num_tokens": 4149885.0,
      "step": 1810
    },
    {
      "entropy": 6.183619403839112,
      "epoch": 0.17941874258600238,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999266541453283,
      "loss": 6.2727,
      "mean_token_accuracy": 0.1341182231903076,
      "num_tokens": 4161824.0,
      "step": 1815
    },
    {
      "entropy": 6.192480278015137,
      "epoch": 0.17991300909450375,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004999257503729487,
      "loss": 6.2305,
      "mean_token_accuracy": 0.13566601723432542,
      "num_tokens": 4173581.0,
      "step": 1820
    },
    {
      "entropy": 6.188820171356201,
      "epoch": 0.18040727560300515,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999248410673607,
      "loss": 6.1086,
      "mean_token_accuracy": 0.14354413226246834,
      "num_tokens": 4184843.0,
      "step": 1825
    },
    {
      "entropy": 6.199634456634522,
      "epoch": 0.18090154211150652,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999239262285869,
      "loss": 6.196,
      "mean_token_accuracy": 0.12894793301820756,
      "num_tokens": 4195016.0,
      "step": 1830
    },
    {
      "entropy": 6.183595895767212,
      "epoch": 0.1813958086200079,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004999230058566495,
      "loss": 6.149,
      "mean_token_accuracy": 0.13868849724531174,
      "num_tokens": 4206798.0,
      "step": 1835
    },
    {
      "entropy": 6.20193657875061,
      "epoch": 0.1818900751285093,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004999220799515713,
      "loss": 6.1197,
      "mean_token_accuracy": 0.13616999015212058,
      "num_tokens": 4218784.0,
      "step": 1840
    },
    {
      "entropy": 6.098097658157348,
      "epoch": 0.18238434163701067,
      "grad_norm": 0.96875,
      "learning_rate": 0.000499921148513375,
      "loss": 6.1495,
      "mean_token_accuracy": 0.13528309762477875,
      "num_tokens": 4229875.0,
      "step": 1845
    },
    {
      "entropy": 6.120871686935425,
      "epoch": 0.18287860814551207,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004999202115420836,
      "loss": 6.1274,
      "mean_token_accuracy": 0.13959604278206825,
      "num_tokens": 4241801.0,
      "step": 1850
    },
    {
      "entropy": 6.136850357055664,
      "epoch": 0.18337287465401345,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999192690377201,
      "loss": 6.112,
      "mean_token_accuracy": 0.13544971346855164,
      "num_tokens": 4253168.0,
      "step": 1855
    },
    {
      "entropy": 6.151134204864502,
      "epoch": 0.18386714116251482,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004999183210003076,
      "loss": 6.1371,
      "mean_token_accuracy": 0.13654277250170707,
      "num_tokens": 4264791.0,
      "step": 1860
    },
    {
      "entropy": 6.127550983428955,
      "epoch": 0.18436140767101622,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999173674298696,
      "loss": 6.0802,
      "mean_token_accuracy": 0.1429269403219223,
      "num_tokens": 4275816.0,
      "step": 1865
    },
    {
      "entropy": 6.2200188636779785,
      "epoch": 0.1848556741795176,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999164083264294,
      "loss": 6.1638,
      "mean_token_accuracy": 0.13557607755064965,
      "num_tokens": 4287806.0,
      "step": 1870
    },
    {
      "entropy": 6.1365429878234865,
      "epoch": 0.185349940688019,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004999154436900108,
      "loss": 6.2061,
      "mean_token_accuracy": 0.13776358291506768,
      "num_tokens": 4300319.0,
      "step": 1875
    },
    {
      "entropy": 6.261188077926636,
      "epoch": 0.18584420719652037,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004999144735206373,
      "loss": 6.2191,
      "mean_token_accuracy": 0.13050539270043374,
      "num_tokens": 4310987.0,
      "step": 1880
    },
    {
      "entropy": 6.134886789321899,
      "epoch": 0.18633847370502174,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004999134978183328,
      "loss": 6.1589,
      "mean_token_accuracy": 0.14048967361450196,
      "num_tokens": 4322438.0,
      "step": 1885
    },
    {
      "entropy": 6.216308641433716,
      "epoch": 0.18683274021352314,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999125165831215,
      "loss": 6.194,
      "mean_token_accuracy": 0.13951309472322465,
      "num_tokens": 4333242.0,
      "step": 1890
    },
    {
      "entropy": 6.1126538753509525,
      "epoch": 0.1873270067220245,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999115298150273,
      "loss": 6.0433,
      "mean_token_accuracy": 0.14102249965071678,
      "num_tokens": 4343660.0,
      "step": 1895
    },
    {
      "entropy": 6.163443613052368,
      "epoch": 0.1878212732305259,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999105375140746,
      "loss": 6.1246,
      "mean_token_accuracy": 0.1411570593714714,
      "num_tokens": 4353382.0,
      "step": 1900
    },
    {
      "entropy": 6.103315830230713,
      "epoch": 0.1883155397390273,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999095396802878,
      "loss": 6.1044,
      "mean_token_accuracy": 0.13109617531299592,
      "num_tokens": 4364614.0,
      "step": 1905
    },
    {
      "entropy": 6.2517712116241455,
      "epoch": 0.18880980624752866,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999085363136913,
      "loss": 6.1812,
      "mean_token_accuracy": 0.1353604979813099,
      "num_tokens": 4376001.0,
      "step": 1910
    },
    {
      "entropy": 6.128733491897583,
      "epoch": 0.18930407275603006,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00049990752741431,
      "loss": 6.1848,
      "mean_token_accuracy": 0.13217584118247033,
      "num_tokens": 4387083.0,
      "step": 1915
    },
    {
      "entropy": 6.139519214630127,
      "epoch": 0.18979833926453143,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004999065129821687,
      "loss": 6.1356,
      "mean_token_accuracy": 0.13750590831041337,
      "num_tokens": 4399977.0,
      "step": 1920
    },
    {
      "entropy": 6.0825254917144775,
      "epoch": 0.1902926057730328,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004999054930172923,
      "loss": 6.0592,
      "mean_token_accuracy": 0.14218987300992011,
      "num_tokens": 4411609.0,
      "step": 1925
    },
    {
      "entropy": 6.175016450881958,
      "epoch": 0.1907868722815342,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999044675197058,
      "loss": 6.1817,
      "mean_token_accuracy": 0.13701564595103263,
      "num_tokens": 4423567.0,
      "step": 1930
    },
    {
      "entropy": 6.147806310653687,
      "epoch": 0.19128113879003558,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999034364894345,
      "loss": 6.1472,
      "mean_token_accuracy": 0.13876281678676605,
      "num_tokens": 4435410.0,
      "step": 1935
    },
    {
      "entropy": 6.18219690322876,
      "epoch": 0.19177540529853698,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999023999265039,
      "loss": 6.1697,
      "mean_token_accuracy": 0.1378607526421547,
      "num_tokens": 4445279.0,
      "step": 1940
    },
    {
      "entropy": 6.077285814285278,
      "epoch": 0.19226967180703836,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999013578309392,
      "loss": 6.0385,
      "mean_token_accuracy": 0.14334971085190773,
      "num_tokens": 4456017.0,
      "step": 1945
    },
    {
      "entropy": 6.126475811004639,
      "epoch": 0.19276393831553973,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999003102027663,
      "loss": 6.0311,
      "mean_token_accuracy": 0.14338691681623458,
      "num_tokens": 4466759.0,
      "step": 1950
    },
    {
      "entropy": 6.117775201797485,
      "epoch": 0.19325820482404113,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000499899257042011,
      "loss": 6.1139,
      "mean_token_accuracy": 0.1352181151509285,
      "num_tokens": 4478316.0,
      "step": 1955
    },
    {
      "entropy": 6.208990001678467,
      "epoch": 0.1937524713325425,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004998981983486988,
      "loss": 6.1751,
      "mean_token_accuracy": 0.1322692520916462,
      "num_tokens": 4490775.0,
      "step": 1960
    },
    {
      "entropy": 6.064915752410888,
      "epoch": 0.1942467378410439,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004998971341228563,
      "loss": 6.039,
      "mean_token_accuracy": 0.1443925455212593,
      "num_tokens": 4501622.0,
      "step": 1965
    },
    {
      "entropy": 6.149889087677002,
      "epoch": 0.19474100434954528,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004998960643645093,
      "loss": 6.1338,
      "mean_token_accuracy": 0.13769015669822693,
      "num_tokens": 4512279.0,
      "step": 1970
    },
    {
      "entropy": 6.167524814605713,
      "epoch": 0.19523527085804665,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004998949890736841,
      "loss": 6.1239,
      "mean_token_accuracy": 0.14070178642868997,
      "num_tokens": 4524087.0,
      "step": 1975
    },
    {
      "entropy": 6.139242649078369,
      "epoch": 0.19572953736654805,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004998939082504075,
      "loss": 6.1075,
      "mean_token_accuracy": 0.13732974529266356,
      "num_tokens": 4534816.0,
      "step": 1980
    },
    {
      "entropy": 6.103002882003784,
      "epoch": 0.19622380387504942,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004998928218947057,
      "loss": 6.0825,
      "mean_token_accuracy": 0.13870148733258247,
      "num_tokens": 4545725.0,
      "step": 1985
    },
    {
      "entropy": 6.218700933456421,
      "epoch": 0.1967180703835508,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004998917300066056,
      "loss": 6.1897,
      "mean_token_accuracy": 0.13517609983682632,
      "num_tokens": 4557003.0,
      "step": 1990
    },
    {
      "entropy": 6.099284267425537,
      "epoch": 0.1972123368920522,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004998906325861342,
      "loss": 6.1023,
      "mean_token_accuracy": 0.13518477007746696,
      "num_tokens": 4568768.0,
      "step": 1995
    },
    {
      "entropy": 6.103130531311035,
      "epoch": 0.19770660340055357,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004998895296333183,
      "loss": 6.0715,
      "mean_token_accuracy": 0.14329598397016524,
      "num_tokens": 4581116.0,
      "step": 2000
    },
    {
      "entropy": 6.170775890350342,
      "epoch": 0.19820086990905497,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000499888421148185,
      "loss": 6.1516,
      "mean_token_accuracy": 0.13928177580237389,
      "num_tokens": 4592660.0,
      "step": 2005
    },
    {
      "entropy": 6.074905729293823,
      "epoch": 0.19869513641755635,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004998873071307617,
      "loss": 6.1321,
      "mean_token_accuracy": 0.13926766365766524,
      "num_tokens": 4603057.0,
      "step": 2010
    },
    {
      "entropy": 6.111081838607788,
      "epoch": 0.19918940292605772,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004998861875810758,
      "loss": 6.0145,
      "mean_token_accuracy": 0.14594943076372147,
      "num_tokens": 4613546.0,
      "step": 2015
    },
    {
      "entropy": 6.130213642120362,
      "epoch": 0.19968366943455912,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004998850624991548,
      "loss": 6.0932,
      "mean_token_accuracy": 0.13966753333806992,
      "num_tokens": 4624613.0,
      "step": 2020
    },
    {
      "entropy": 6.070725202560425,
      "epoch": 0.2001779359430605,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004998839318850263,
      "loss": 6.0581,
      "mean_token_accuracy": 0.14254929646849632,
      "num_tokens": 4636004.0,
      "step": 2025
    },
    {
      "entropy": 6.094811868667603,
      "epoch": 0.2006722024515619,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998827957387182,
      "loss": 6.1194,
      "mean_token_accuracy": 0.1468837432563305,
      "num_tokens": 4646450.0,
      "step": 2030
    },
    {
      "entropy": 6.105382490158081,
      "epoch": 0.20116646896006327,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004998816540602584,
      "loss": 6.0993,
      "mean_token_accuracy": 0.14319370239973067,
      "num_tokens": 4658928.0,
      "step": 2035
    },
    {
      "entropy": 6.133113861083984,
      "epoch": 0.20166073546856464,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004998805068496752,
      "loss": 6.1279,
      "mean_token_accuracy": 0.13475392609834672,
      "num_tokens": 4670392.0,
      "step": 2040
    },
    {
      "entropy": 6.141398954391479,
      "epoch": 0.20215500197706604,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004998793541069964,
      "loss": 6.108,
      "mean_token_accuracy": 0.13688452690839767,
      "num_tokens": 4680605.0,
      "step": 2045
    },
    {
      "entropy": 6.069099378585816,
      "epoch": 0.2026492684855674,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004998781958322507,
      "loss": 6.0134,
      "mean_token_accuracy": 0.14460673555731773,
      "num_tokens": 4694250.0,
      "step": 2050
    },
    {
      "entropy": 6.110835599899292,
      "epoch": 0.20314353499406881,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004998770320254666,
      "loss": 6.1661,
      "mean_token_accuracy": 0.13903781324625014,
      "num_tokens": 4705754.0,
      "step": 2055
    },
    {
      "entropy": 6.213265514373779,
      "epoch": 0.2036378015025702,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004998758626866725,
      "loss": 6.0439,
      "mean_token_accuracy": 0.14397211521863937,
      "num_tokens": 4716393.0,
      "step": 2060
    },
    {
      "entropy": 5.992143011093139,
      "epoch": 0.20413206801107156,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004998746878158973,
      "loss": 6.055,
      "mean_token_accuracy": 0.14380352646112443,
      "num_tokens": 4726894.0,
      "step": 2065
    },
    {
      "entropy": 6.137889289855957,
      "epoch": 0.20462633451957296,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998735074131699,
      "loss": 6.082,
      "mean_token_accuracy": 0.13775463551282882,
      "num_tokens": 4737488.0,
      "step": 2070
    },
    {
      "entropy": 6.157661390304566,
      "epoch": 0.20512060102807433,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004998723214785194,
      "loss": 6.1528,
      "mean_token_accuracy": 0.1327391304075718,
      "num_tokens": 4749131.0,
      "step": 2075
    },
    {
      "entropy": 6.090861511230469,
      "epoch": 0.20561486753657574,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004998711300119749,
      "loss": 6.0322,
      "mean_token_accuracy": 0.1410945810377598,
      "num_tokens": 4760503.0,
      "step": 2080
    },
    {
      "entropy": 6.128068828582764,
      "epoch": 0.2061091340450771,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004998699330135656,
      "loss": 6.1567,
      "mean_token_accuracy": 0.1358460932970047,
      "num_tokens": 4771965.0,
      "step": 2085
    },
    {
      "entropy": 6.096466827392578,
      "epoch": 0.20660340055357848,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004998687304833212,
      "loss": 6.1254,
      "mean_token_accuracy": 0.13651231974363326,
      "num_tokens": 4784215.0,
      "step": 2090
    },
    {
      "entropy": 6.102608966827392,
      "epoch": 0.20709766706207988,
      "grad_norm": 1.078125,
      "learning_rate": 0.000499867522421271,
      "loss": 6.0457,
      "mean_token_accuracy": 0.1427777349948883,
      "num_tokens": 4796410.0,
      "step": 2095
    },
    {
      "entropy": 6.065385389328003,
      "epoch": 0.20759193357058126,
      "grad_norm": 1.109375,
      "learning_rate": 0.000499866308827445,
      "loss": 6.0463,
      "mean_token_accuracy": 0.1456620767712593,
      "num_tokens": 4806990.0,
      "step": 2100
    },
    {
      "entropy": 6.113207626342773,
      "epoch": 0.20808620007908263,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004998650897018729,
      "loss": 6.0949,
      "mean_token_accuracy": 0.14114938974380492,
      "num_tokens": 4818903.0,
      "step": 2105
    },
    {
      "entropy": 6.091670751571655,
      "epoch": 0.20858046658758403,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004998638650445847,
      "loss": 6.0222,
      "mean_token_accuracy": 0.14184736907482148,
      "num_tokens": 4829130.0,
      "step": 2110
    },
    {
      "entropy": 6.0775425910949705,
      "epoch": 0.2090747330960854,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004998626348556105,
      "loss": 6.0507,
      "mean_token_accuracy": 0.1536749005317688,
      "num_tokens": 4840424.0,
      "step": 2115
    },
    {
      "entropy": 6.098685884475708,
      "epoch": 0.2095689996045868,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004998613991349807,
      "loss": 6.0283,
      "mean_token_accuracy": 0.143294495344162,
      "num_tokens": 4851674.0,
      "step": 2120
    },
    {
      "entropy": 5.9475749969482425,
      "epoch": 0.21006326611308818,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004998601578827255,
      "loss": 5.9793,
      "mean_token_accuracy": 0.1536559596657753,
      "num_tokens": 4863884.0,
      "step": 2125
    },
    {
      "entropy": 6.033754873275757,
      "epoch": 0.21055753262158955,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004998589110988755,
      "loss": 6.0494,
      "mean_token_accuracy": 0.14718157574534416,
      "num_tokens": 4875582.0,
      "step": 2130
    },
    {
      "entropy": 6.114961338043213,
      "epoch": 0.21105179913009095,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004998576587834614,
      "loss": 6.0803,
      "mean_token_accuracy": 0.14199472591280937,
      "num_tokens": 4888325.0,
      "step": 2135
    },
    {
      "entropy": 6.098699426651001,
      "epoch": 0.21154606563859232,
      "grad_norm": 1.046875,
      "learning_rate": 0.000499856400936514,
      "loss": 6.072,
      "mean_token_accuracy": 0.14264639914035798,
      "num_tokens": 4899631.0,
      "step": 2140
    },
    {
      "entropy": 6.029827070236206,
      "epoch": 0.21204033214709372,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004998551375580643,
      "loss": 6.0712,
      "mean_token_accuracy": 0.13763367459177972,
      "num_tokens": 4910954.0,
      "step": 2145
    },
    {
      "entropy": 6.051994514465332,
      "epoch": 0.2125345986555951,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004998538686481435,
      "loss": 6.0714,
      "mean_token_accuracy": 0.14147689938545227,
      "num_tokens": 4921952.0,
      "step": 2150
    },
    {
      "entropy": 6.0656555652618405,
      "epoch": 0.21302886516409647,
      "grad_norm": 1.0,
      "learning_rate": 0.0004998525942067824,
      "loss": 6.0253,
      "mean_token_accuracy": 0.14360493123531343,
      "num_tokens": 4933984.0,
      "step": 2155
    },
    {
      "entropy": 6.164171361923218,
      "epoch": 0.21352313167259787,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004998513142340128,
      "loss": 6.1657,
      "mean_token_accuracy": 0.14157587438821792,
      "num_tokens": 4945926.0,
      "step": 2160
    },
    {
      "entropy": 6.026093053817749,
      "epoch": 0.21401739818109924,
      "grad_norm": 1.125,
      "learning_rate": 0.0004998500287298657,
      "loss": 6.0878,
      "mean_token_accuracy": 0.14866259396076204,
      "num_tokens": 4958123.0,
      "step": 2165
    },
    {
      "entropy": 6.158975839614868,
      "epoch": 0.21451166468960065,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004998487376943733,
      "loss": 6.1491,
      "mean_token_accuracy": 0.13338658437132836,
      "num_tokens": 4970409.0,
      "step": 2170
    },
    {
      "entropy": 6.122834396362305,
      "epoch": 0.21500593119810202,
      "grad_norm": 1.09375,
      "learning_rate": 0.000499847441127567,
      "loss": 6.0485,
      "mean_token_accuracy": 0.13795322477817534,
      "num_tokens": 4980429.0,
      "step": 2175
    },
    {
      "entropy": 6.053185510635376,
      "epoch": 0.2155001977066034,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998461390294787,
      "loss": 6.0711,
      "mean_token_accuracy": 0.14336730688810348,
      "num_tokens": 4991790.0,
      "step": 2180
    },
    {
      "entropy": 6.095970296859742,
      "epoch": 0.2159944642151048,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004998448314001405,
      "loss": 6.0869,
      "mean_token_accuracy": 0.14493904113769532,
      "num_tokens": 5003611.0,
      "step": 2185
    },
    {
      "entropy": 6.176206493377686,
      "epoch": 0.21648873072360617,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004998435182395845,
      "loss": 6.1385,
      "mean_token_accuracy": 0.1309673383831978,
      "num_tokens": 5014824.0,
      "step": 2190
    },
    {
      "entropy": 6.142057228088379,
      "epoch": 0.21698299723210757,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004998421995478432,
      "loss": 6.0831,
      "mean_token_accuracy": 0.143635593354702,
      "num_tokens": 5026937.0,
      "step": 2195
    },
    {
      "entropy": 6.116099071502686,
      "epoch": 0.21747726374060894,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004998408753249489,
      "loss": 6.1134,
      "mean_token_accuracy": 0.1369356207549572,
      "num_tokens": 5037996.0,
      "step": 2200
    },
    {
      "entropy": 6.096542501449585,
      "epoch": 0.2179715302491103,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004998395455709342,
      "loss": 6.0605,
      "mean_token_accuracy": 0.14553544148802758,
      "num_tokens": 5050944.0,
      "step": 2205
    },
    {
      "entropy": 6.150865888595581,
      "epoch": 0.2184657967576117,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004998382102858317,
      "loss": 6.1,
      "mean_token_accuracy": 0.1378478489816189,
      "num_tokens": 5062129.0,
      "step": 2210
    },
    {
      "entropy": 6.0926353454589846,
      "epoch": 0.2189600632661131,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004998368694696743,
      "loss": 6.0959,
      "mean_token_accuracy": 0.14366044774651526,
      "num_tokens": 5073318.0,
      "step": 2215
    },
    {
      "entropy": 6.064820623397827,
      "epoch": 0.21945432977461446,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004998355231224952,
      "loss": 6.1278,
      "mean_token_accuracy": 0.13671059682965278,
      "num_tokens": 5084607.0,
      "step": 2220
    },
    {
      "entropy": 6.1097174167633055,
      "epoch": 0.21994859628311586,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004998341712443273,
      "loss": 6.0165,
      "mean_token_accuracy": 0.14424162060022355,
      "num_tokens": 5096004.0,
      "step": 2225
    },
    {
      "entropy": 6.0761024951934814,
      "epoch": 0.22044286279161723,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000499832813835204,
      "loss": 6.0296,
      "mean_token_accuracy": 0.14155133366584777,
      "num_tokens": 5107171.0,
      "step": 2230
    },
    {
      "entropy": 5.977806282043457,
      "epoch": 0.22093712930011863,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004998314508951584,
      "loss": 5.9782,
      "mean_token_accuracy": 0.15077303797006608,
      "num_tokens": 5119249.0,
      "step": 2235
    },
    {
      "entropy": 6.04344334602356,
      "epoch": 0.22143139580862,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004998300824242244,
      "loss": 5.9945,
      "mean_token_accuracy": 0.13899804279208183,
      "num_tokens": 5130798.0,
      "step": 2240
    },
    {
      "entropy": 6.146693372726441,
      "epoch": 0.22192566231712138,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004998287084224353,
      "loss": 6.1624,
      "mean_token_accuracy": 0.12782063856720924,
      "num_tokens": 5141376.0,
      "step": 2245
    },
    {
      "entropy": 5.979104852676391,
      "epoch": 0.22241992882562278,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004998273288898252,
      "loss": 5.9599,
      "mean_token_accuracy": 0.1464287005364895,
      "num_tokens": 5152888.0,
      "step": 2250
    },
    {
      "entropy": 6.031632804870606,
      "epoch": 0.22291419533412415,
      "grad_norm": 0.9375,
      "learning_rate": 0.000499825943826428,
      "loss": 6.0647,
      "mean_token_accuracy": 0.14004838690161706,
      "num_tokens": 5165720.0,
      "step": 2255
    },
    {
      "entropy": 6.069583225250244,
      "epoch": 0.22340846184262556,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004998245532322776,
      "loss": 5.9363,
      "mean_token_accuracy": 0.14667778834700584,
      "num_tokens": 5177345.0,
      "step": 2260
    },
    {
      "entropy": 6.060588693618774,
      "epoch": 0.22390272835112693,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004998231571074083,
      "loss": 5.9634,
      "mean_token_accuracy": 0.14570290222764015,
      "num_tokens": 5187385.0,
      "step": 2265
    },
    {
      "entropy": 6.04728593826294,
      "epoch": 0.2243969948596283,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004998217554518545,
      "loss": 6.0683,
      "mean_token_accuracy": 0.13824618086218834,
      "num_tokens": 5199552.0,
      "step": 2270
    },
    {
      "entropy": 6.039723873138428,
      "epoch": 0.2248912613681297,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004998203482656507,
      "loss": 6.027,
      "mean_token_accuracy": 0.14244999885559081,
      "num_tokens": 5210316.0,
      "step": 2275
    },
    {
      "entropy": 6.0842948913574215,
      "epoch": 0.22538552787663108,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004998189355488314,
      "loss": 6.1053,
      "mean_token_accuracy": 0.1472305215895176,
      "num_tokens": 5221595.0,
      "step": 2280
    },
    {
      "entropy": 6.140758895874024,
      "epoch": 0.22587979438513248,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998175173014313,
      "loss": 6.0722,
      "mean_token_accuracy": 0.13632547855377197,
      "num_tokens": 5232882.0,
      "step": 2285
    },
    {
      "entropy": 6.0675756454467775,
      "epoch": 0.22637406089363385,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998160935234857,
      "loss": 6.1709,
      "mean_token_accuracy": 0.13839787468314171,
      "num_tokens": 5244841.0,
      "step": 2290
    },
    {
      "entropy": 6.096714353561401,
      "epoch": 0.22686832740213522,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000499814664215029,
      "loss": 5.9748,
      "mean_token_accuracy": 0.1458760157227516,
      "num_tokens": 5256162.0,
      "step": 2295
    },
    {
      "entropy": 6.00123872756958,
      "epoch": 0.22736259391063662,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004998132293760968,
      "loss": 5.9029,
      "mean_token_accuracy": 0.15427679419517518,
      "num_tokens": 5267130.0,
      "step": 2300
    },
    {
      "entropy": 5.969837474822998,
      "epoch": 0.227856860419138,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004998117890067243,
      "loss": 6.0244,
      "mean_token_accuracy": 0.14411933422088624,
      "num_tokens": 5279394.0,
      "step": 2305
    },
    {
      "entropy": 5.981906747817993,
      "epoch": 0.22835112692763937,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004998103431069469,
      "loss": 5.9975,
      "mean_token_accuracy": 0.1491586372256279,
      "num_tokens": 5291068.0,
      "step": 2310
    },
    {
      "entropy": 6.060888195037842,
      "epoch": 0.22884539343614077,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004998088916768002,
      "loss": 5.9579,
      "mean_token_accuracy": 0.14889078661799432,
      "num_tokens": 5300924.0,
      "step": 2315
    },
    {
      "entropy": 6.016797971725464,
      "epoch": 0.22933965994464214,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004998074347163199,
      "loss": 5.9542,
      "mean_token_accuracy": 0.1480551391839981,
      "num_tokens": 5311998.0,
      "step": 2320
    },
    {
      "entropy": 6.06162691116333,
      "epoch": 0.22983392645314354,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004998059722255418,
      "loss": 5.9899,
      "mean_token_accuracy": 0.14691204354166984,
      "num_tokens": 5322163.0,
      "step": 2325
    },
    {
      "entropy": 6.0275694847106935,
      "epoch": 0.23032819296164492,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004998045042045019,
      "loss": 5.996,
      "mean_token_accuracy": 0.1497631698846817,
      "num_tokens": 5331638.0,
      "step": 2330
    },
    {
      "entropy": 6.048976612091065,
      "epoch": 0.2308224594701463,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998030306532363,
      "loss": 5.9939,
      "mean_token_accuracy": 0.14958565384149553,
      "num_tokens": 5342399.0,
      "step": 2335
    },
    {
      "entropy": 6.055438804626465,
      "epoch": 0.2313167259786477,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004998015515717813,
      "loss": 6.0695,
      "mean_token_accuracy": 0.14016990289092063,
      "num_tokens": 5353481.0,
      "step": 2340
    },
    {
      "entropy": 6.052402687072754,
      "epoch": 0.23181099248714906,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998000669601733,
      "loss": 5.9689,
      "mean_token_accuracy": 0.1473512254655361,
      "num_tokens": 5365167.0,
      "step": 2345
    },
    {
      "entropy": 6.002137184143066,
      "epoch": 0.23230525899565047,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004997985768184488,
      "loss": 5.9967,
      "mean_token_accuracy": 0.1431095339357853,
      "num_tokens": 5375809.0,
      "step": 2350
    },
    {
      "entropy": 5.962526512145996,
      "epoch": 0.23279952550415184,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004997970811466443,
      "loss": 5.9133,
      "mean_token_accuracy": 0.1538688361644745,
      "num_tokens": 5387372.0,
      "step": 2355
    },
    {
      "entropy": 6.187628889083863,
      "epoch": 0.2332937920126532,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004997955799447969,
      "loss": 6.0898,
      "mean_token_accuracy": 0.13684846088290215,
      "num_tokens": 5399210.0,
      "step": 2360
    },
    {
      "entropy": 5.985762119293213,
      "epoch": 0.2337880585211546,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004997940732129432,
      "loss": 6.017,
      "mean_token_accuracy": 0.14093376472592353,
      "num_tokens": 5410969.0,
      "step": 2365
    },
    {
      "entropy": 5.984814739227295,
      "epoch": 0.23428232502965599,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004997925609511205,
      "loss": 5.9824,
      "mean_token_accuracy": 0.14780228435993195,
      "num_tokens": 5422569.0,
      "step": 2370
    },
    {
      "entropy": 6.031500720977784,
      "epoch": 0.2347765915381574,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004997910431593659,
      "loss": 5.9606,
      "mean_token_accuracy": 0.15011014714837073,
      "num_tokens": 5434592.0,
      "step": 2375
    },
    {
      "entropy": 6.0243175506591795,
      "epoch": 0.23527085804665876,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004997895198377168,
      "loss": 6.0545,
      "mean_token_accuracy": 0.14284207075834274,
      "num_tokens": 5446525.0,
      "step": 2380
    },
    {
      "entropy": 6.08224196434021,
      "epoch": 0.23576512455516013,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004997879909862105,
      "loss": 6.025,
      "mean_token_accuracy": 0.14782738387584687,
      "num_tokens": 5457327.0,
      "step": 2385
    },
    {
      "entropy": 6.105248498916626,
      "epoch": 0.23625939106366153,
      "grad_norm": 1.078125,
      "learning_rate": 0.000499786456604885,
      "loss": 6.0642,
      "mean_token_accuracy": 0.1362588956952095,
      "num_tokens": 5468533.0,
      "step": 2390
    },
    {
      "entropy": 5.998384857177735,
      "epoch": 0.2367536575721629,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004997849166937776,
      "loss": 5.9451,
      "mean_token_accuracy": 0.14417133554816247,
      "num_tokens": 5480652.0,
      "step": 2395
    },
    {
      "entropy": 6.012068605422973,
      "epoch": 0.2372479240806643,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004997833712529265,
      "loss": 5.9707,
      "mean_token_accuracy": 0.14451036602258682,
      "num_tokens": 5491446.0,
      "step": 2400
    },
    {
      "entropy": 6.021755743026733,
      "epoch": 0.23774219058916568,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004997818202823695,
      "loss": 5.9945,
      "mean_token_accuracy": 0.14959129244089125,
      "num_tokens": 5502886.0,
      "step": 2405
    },
    {
      "entropy": 6.0749561309814455,
      "epoch": 0.23823645709766705,
      "grad_norm": 1.125,
      "learning_rate": 0.0004997802637821449,
      "loss": 6.023,
      "mean_token_accuracy": 0.14053252562880517,
      "num_tokens": 5514305.0,
      "step": 2410
    },
    {
      "entropy": 5.98513970375061,
      "epoch": 0.23873072360616845,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004997787017522909,
      "loss": 5.9437,
      "mean_token_accuracy": 0.15076548904180526,
      "num_tokens": 5525913.0,
      "step": 2415
    },
    {
      "entropy": 5.90565128326416,
      "epoch": 0.23922499011466983,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004997771341928459,
      "loss": 5.9446,
      "mean_token_accuracy": 0.1544351950287819,
      "num_tokens": 5537069.0,
      "step": 2420
    },
    {
      "entropy": 6.022730112075806,
      "epoch": 0.2397192566231712,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004997755611038487,
      "loss": 6.0441,
      "mean_token_accuracy": 0.1400386117398739,
      "num_tokens": 5547988.0,
      "step": 2425
    },
    {
      "entropy": 6.105308675765992,
      "epoch": 0.2402135231316726,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004997739824853378,
      "loss": 6.1088,
      "mean_token_accuracy": 0.1395256355404854,
      "num_tokens": 5560400.0,
      "step": 2430
    },
    {
      "entropy": 5.987665414810181,
      "epoch": 0.24070778964017397,
      "grad_norm": 1.0,
      "learning_rate": 0.0004997723983373519,
      "loss": 5.9696,
      "mean_token_accuracy": 0.15029642432928086,
      "num_tokens": 5572291.0,
      "step": 2435
    },
    {
      "entropy": 6.100761747360229,
      "epoch": 0.24120205614867538,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004997708086599303,
      "loss": 5.9938,
      "mean_token_accuracy": 0.14239495247602463,
      "num_tokens": 5583583.0,
      "step": 2440
    },
    {
      "entropy": 6.093548583984375,
      "epoch": 0.24169632265717675,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004997692134531119,
      "loss": 6.0205,
      "mean_token_accuracy": 0.14349138885736465,
      "num_tokens": 5595499.0,
      "step": 2445
    },
    {
      "entropy": 5.937535381317138,
      "epoch": 0.24219058916567812,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000499767612716936,
      "loss": 5.9486,
      "mean_token_accuracy": 0.1555730566382408,
      "num_tokens": 5606745.0,
      "step": 2450
    },
    {
      "entropy": 5.977556610107422,
      "epoch": 0.24268485567417952,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004997660064514419,
      "loss": 6.0333,
      "mean_token_accuracy": 0.1459631770849228,
      "num_tokens": 5618226.0,
      "step": 2455
    },
    {
      "entropy": 6.103812122344971,
      "epoch": 0.2431791221826809,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004997643946566691,
      "loss": 6.0358,
      "mean_token_accuracy": 0.14382674545049667,
      "num_tokens": 5628941.0,
      "step": 2460
    },
    {
      "entropy": 5.948967981338501,
      "epoch": 0.2436733886911823,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004997627773326574,
      "loss": 5.8499,
      "mean_token_accuracy": 0.15501064658164979,
      "num_tokens": 5641077.0,
      "step": 2465
    },
    {
      "entropy": 6.018094301223755,
      "epoch": 0.24416765519968367,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004997611544794465,
      "loss": 6.1555,
      "mean_token_accuracy": 0.1338454492390156,
      "num_tokens": 5652156.0,
      "step": 2470
    },
    {
      "entropy": 6.151158618927002,
      "epoch": 0.24466192170818504,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004997595260970764,
      "loss": 6.0822,
      "mean_token_accuracy": 0.14347486943006516,
      "num_tokens": 5663988.0,
      "step": 2475
    },
    {
      "entropy": 5.977680206298828,
      "epoch": 0.24515618821668644,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000499757892185587,
      "loss": 6.0018,
      "mean_token_accuracy": 0.14267092123627662,
      "num_tokens": 5676302.0,
      "step": 2480
    },
    {
      "entropy": 6.070641231536865,
      "epoch": 0.24565045472518782,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004997562527450187,
      "loss": 6.046,
      "mean_token_accuracy": 0.14550234749913216,
      "num_tokens": 5686686.0,
      "step": 2485
    },
    {
      "entropy": 6.021747350692749,
      "epoch": 0.24614472123368922,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004997546077754116,
      "loss": 5.9525,
      "mean_token_accuracy": 0.15929894894361496,
      "num_tokens": 5698379.0,
      "step": 2490
    },
    {
      "entropy": 5.958306932449341,
      "epoch": 0.2466389877421906,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004997529572768063,
      "loss": 5.9432,
      "mean_token_accuracy": 0.1498410865664482,
      "num_tokens": 5708891.0,
      "step": 2495
    },
    {
      "entropy": 5.981058549880982,
      "epoch": 0.24713325425069196,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004997513012492434,
      "loss": 5.9923,
      "mean_token_accuracy": 0.14332532063126563,
      "num_tokens": 5720326.0,
      "step": 2500
    },
    {
      "entropy": 6.052645683288574,
      "epoch": 0.24762752075919336,
      "grad_norm": 1.0,
      "learning_rate": 0.0004997496396927636,
      "loss": 5.9112,
      "mean_token_accuracy": 0.15235294178128242,
      "num_tokens": 5731121.0,
      "step": 2505
    },
    {
      "entropy": 6.023140573501587,
      "epoch": 0.24812178726769474,
      "grad_norm": 1.125,
      "learning_rate": 0.0004997479726074077,
      "loss": 5.9597,
      "mean_token_accuracy": 0.15026197880506514,
      "num_tokens": 5741637.0,
      "step": 2510
    },
    {
      "entropy": 6.048560428619385,
      "epoch": 0.24861605377619614,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000499746299993217,
      "loss": 6.0102,
      "mean_token_accuracy": 0.14581428989768028,
      "num_tokens": 5752659.0,
      "step": 2515
    },
    {
      "entropy": 5.998668670654297,
      "epoch": 0.2491103202846975,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004997446218502324,
      "loss": 5.9728,
      "mean_token_accuracy": 0.14580842703580857,
      "num_tokens": 5763829.0,
      "step": 2520
    },
    {
      "entropy": 6.0419858455657955,
      "epoch": 0.24960458679319888,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004997429381784952,
      "loss": 6.053,
      "mean_token_accuracy": 0.1448736622929573,
      "num_tokens": 5775848.0,
      "step": 2525
    },
    {
      "entropy": 5.984309673309326,
      "epoch": 0.2500988533017003,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004997412489780469,
      "loss": 6.0037,
      "mean_token_accuracy": 0.1499286875128746,
      "num_tokens": 5786595.0,
      "step": 2530
    },
    {
      "entropy": 5.964243793487549,
      "epoch": 0.25059311981020166,
      "grad_norm": 1.0625,
      "learning_rate": 0.000499739554248929,
      "loss": 5.8929,
      "mean_token_accuracy": 0.15398191586136817,
      "num_tokens": 5797841.0,
      "step": 2535
    },
    {
      "entropy": 5.912666082382202,
      "epoch": 0.25108738631870303,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004997378539911832,
      "loss": 5.9089,
      "mean_token_accuracy": 0.15088940635323525,
      "num_tokens": 5809106.0,
      "step": 2540
    },
    {
      "entropy": 5.973687314987183,
      "epoch": 0.2515816528272044,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004997361482048513,
      "loss": 5.9215,
      "mean_token_accuracy": 0.1497866354882717,
      "num_tokens": 5820316.0,
      "step": 2545
    },
    {
      "entropy": 6.026650857925415,
      "epoch": 0.25207591933570583,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004997344368899753,
      "loss": 6.0408,
      "mean_token_accuracy": 0.139362945407629,
      "num_tokens": 5833137.0,
      "step": 2550
    },
    {
      "entropy": 6.037950611114502,
      "epoch": 0.2525701858442072,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004997327200465972,
      "loss": 5.9694,
      "mean_token_accuracy": 0.14529137313365936,
      "num_tokens": 5844088.0,
      "step": 2555
    },
    {
      "entropy": 6.077861881256103,
      "epoch": 0.2530644523527086,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004997309976747594,
      "loss": 5.9736,
      "mean_token_accuracy": 0.14326793625950812,
      "num_tokens": 5854546.0,
      "step": 2560
    },
    {
      "entropy": 6.011197853088379,
      "epoch": 0.25355871886120995,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004997292697745042,
      "loss": 5.9723,
      "mean_token_accuracy": 0.1466481104493141,
      "num_tokens": 5865326.0,
      "step": 2565
    },
    {
      "entropy": 5.999698734283447,
      "epoch": 0.2540529853697113,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004997275363458741,
      "loss": 5.8594,
      "mean_token_accuracy": 0.1503140576183796,
      "num_tokens": 5876487.0,
      "step": 2570
    },
    {
      "entropy": 5.875597381591797,
      "epoch": 0.25454725187821275,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004997257973889118,
      "loss": 5.9726,
      "mean_token_accuracy": 0.14843393564224244,
      "num_tokens": 5887846.0,
      "step": 2575
    },
    {
      "entropy": 6.101366996765137,
      "epoch": 0.25504151838671413,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004997240529036599,
      "loss": 5.9412,
      "mean_token_accuracy": 0.15172709226608277,
      "num_tokens": 5898572.0,
      "step": 2580
    },
    {
      "entropy": 5.881202983856201,
      "epoch": 0.2555357848952155,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004997223028901615,
      "loss": 6.0037,
      "mean_token_accuracy": 0.1475423127412796,
      "num_tokens": 5908927.0,
      "step": 2585
    },
    {
      "entropy": 6.037431478500366,
      "epoch": 0.2560300514037169,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004997205473484596,
      "loss": 5.9495,
      "mean_token_accuracy": 0.1498175874352455,
      "num_tokens": 5919807.0,
      "step": 2590
    },
    {
      "entropy": 5.920142316818238,
      "epoch": 0.25652431791221825,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004997187862785974,
      "loss": 5.9779,
      "mean_token_accuracy": 0.14424620494246482,
      "num_tokens": 5931160.0,
      "step": 2595
    },
    {
      "entropy": 6.088014888763428,
      "epoch": 0.2570185844207197,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004997170196806181,
      "loss": 6.0412,
      "mean_token_accuracy": 0.1426064506173134,
      "num_tokens": 5943694.0,
      "step": 2600
    },
    {
      "entropy": 5.981357526779175,
      "epoch": 0.25751285092922105,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004997152475545652,
      "loss": 5.9319,
      "mean_token_accuracy": 0.14945385083556176,
      "num_tokens": 5954543.0,
      "step": 2605
    },
    {
      "entropy": 5.9636317729949955,
      "epoch": 0.2580071174377224,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004997134699004825,
      "loss": 5.973,
      "mean_token_accuracy": 0.1513287492096424,
      "num_tokens": 5965080.0,
      "step": 2610
    },
    {
      "entropy": 6.036751079559326,
      "epoch": 0.2585013839462238,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004997116867184136,
      "loss": 5.9566,
      "mean_token_accuracy": 0.14630378186702728,
      "num_tokens": 5975314.0,
      "step": 2615
    },
    {
      "entropy": 6.02877459526062,
      "epoch": 0.25899565045472517,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004997098980084022,
      "loss": 5.9735,
      "mean_token_accuracy": 0.14686284810304642,
      "num_tokens": 5985927.0,
      "step": 2620
    },
    {
      "entropy": 5.9730428695678714,
      "epoch": 0.2594899169632266,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004997081037704926,
      "loss": 5.9986,
      "mean_token_accuracy": 0.145076747238636,
      "num_tokens": 5997311.0,
      "step": 2625
    },
    {
      "entropy": 5.986532354354859,
      "epoch": 0.25998418347172797,
      "grad_norm": 1.0,
      "learning_rate": 0.0004997063040047288,
      "loss": 6.0236,
      "mean_token_accuracy": 0.13863110765814782,
      "num_tokens": 6008595.0,
      "step": 2630
    },
    {
      "entropy": 6.084332275390625,
      "epoch": 0.26047844998022934,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004997044987111548,
      "loss": 5.9346,
      "mean_token_accuracy": 0.1485560044646263,
      "num_tokens": 6019055.0,
      "step": 2635
    },
    {
      "entropy": 5.984119462966919,
      "epoch": 0.2609727164887307,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004997026878898155,
      "loss": 6.0785,
      "mean_token_accuracy": 0.13726168423891066,
      "num_tokens": 6030029.0,
      "step": 2640
    },
    {
      "entropy": 6.064934492111206,
      "epoch": 0.2614669829972321,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004997008715407552,
      "loss": 6.0149,
      "mean_token_accuracy": 0.14624494388699533,
      "num_tokens": 6041495.0,
      "step": 2645
    },
    {
      "entropy": 6.0098552227020265,
      "epoch": 0.2619612495057335,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004996990496640185,
      "loss": 5.9223,
      "mean_token_accuracy": 0.14212577491998674,
      "num_tokens": 6052307.0,
      "step": 2650
    },
    {
      "entropy": 5.9654816627502445,
      "epoch": 0.2624555160142349,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004996972222596505,
      "loss": 5.8811,
      "mean_token_accuracy": 0.14910161644220352,
      "num_tokens": 6062550.0,
      "step": 2655
    },
    {
      "entropy": 5.945908880233764,
      "epoch": 0.26294978252273626,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004996953893276958,
      "loss": 5.9301,
      "mean_token_accuracy": 0.14317954927682877,
      "num_tokens": 6074450.0,
      "step": 2660
    },
    {
      "entropy": 5.945242881774902,
      "epoch": 0.26344404903123764,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004996935508681997,
      "loss": 5.9609,
      "mean_token_accuracy": 0.1495286300778389,
      "num_tokens": 6085749.0,
      "step": 2665
    },
    {
      "entropy": 6.005150699615479,
      "epoch": 0.263938315539739,
      "grad_norm": 1.125,
      "learning_rate": 0.0004996917068812075,
      "loss": 5.8905,
      "mean_token_accuracy": 0.1543781578540802,
      "num_tokens": 6096438.0,
      "step": 2670
    },
    {
      "entropy": 5.924571704864502,
      "epoch": 0.26443258204824044,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004996898573667643,
      "loss": 5.9266,
      "mean_token_accuracy": 0.1454748436808586,
      "num_tokens": 6108386.0,
      "step": 2675
    },
    {
      "entropy": 6.008694839477539,
      "epoch": 0.2649268485567418,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004996880023249158,
      "loss": 5.8626,
      "mean_token_accuracy": 0.15368301123380662,
      "num_tokens": 6118771.0,
      "step": 2680
    },
    {
      "entropy": 5.974543571472168,
      "epoch": 0.2654211150652432,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004996861417557076,
      "loss": 5.9496,
      "mean_token_accuracy": 0.14521803110837936,
      "num_tokens": 6130718.0,
      "step": 2685
    },
    {
      "entropy": 5.922163200378418,
      "epoch": 0.26591538157374456,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004996842756591855,
      "loss": 5.9631,
      "mean_token_accuracy": 0.14341218546032905,
      "num_tokens": 6142086.0,
      "step": 2690
    },
    {
      "entropy": 5.976787948608399,
      "epoch": 0.26640964808224593,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004996824040353953,
      "loss": 5.9613,
      "mean_token_accuracy": 0.1457512103021145,
      "num_tokens": 6153857.0,
      "step": 2695
    },
    {
      "entropy": 6.125744962692261,
      "epoch": 0.2669039145907473,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004996805268843832,
      "loss": 6.0951,
      "mean_token_accuracy": 0.1432183265686035,
      "num_tokens": 6166796.0,
      "step": 2700
    },
    {
      "entropy": 5.963466644287109,
      "epoch": 0.26739818109924873,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004996786442061952,
      "loss": 5.9169,
      "mean_token_accuracy": 0.15192901045084,
      "num_tokens": 6178046.0,
      "step": 2705
    },
    {
      "entropy": 5.9657057762146,
      "epoch": 0.2678924476077501,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004996767560008777,
      "loss": 5.9593,
      "mean_token_accuracy": 0.14713699072599412,
      "num_tokens": 6190455.0,
      "step": 2710
    },
    {
      "entropy": 5.899187612533569,
      "epoch": 0.2683867141162515,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004996748622684771,
      "loss": 5.8462,
      "mean_token_accuracy": 0.15217214077711105,
      "num_tokens": 6201659.0,
      "step": 2715
    },
    {
      "entropy": 5.924319171905518,
      "epoch": 0.26888098062475285,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004996729630090401,
      "loss": 5.9552,
      "mean_token_accuracy": 0.1431477539241314,
      "num_tokens": 6213870.0,
      "step": 2720
    },
    {
      "entropy": 6.0065733909606935,
      "epoch": 0.2693752471332542,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004996710582226133,
      "loss": 5.9012,
      "mean_token_accuracy": 0.1548033893108368,
      "num_tokens": 6224237.0,
      "step": 2725
    },
    {
      "entropy": 5.960059928894043,
      "epoch": 0.26986951364175565,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004996691479092437,
      "loss": 5.9433,
      "mean_token_accuracy": 0.14833316206932068,
      "num_tokens": 6235868.0,
      "step": 2730
    },
    {
      "entropy": 6.0247667789459225,
      "epoch": 0.270363780150257,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004996672320689782,
      "loss": 5.9961,
      "mean_token_accuracy": 0.14516259282827376,
      "num_tokens": 6247994.0,
      "step": 2735
    },
    {
      "entropy": 5.855693197250366,
      "epoch": 0.2708580466587584,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004996653107018638,
      "loss": 5.8469,
      "mean_token_accuracy": 0.15463558286428453,
      "num_tokens": 6259654.0,
      "step": 2740
    },
    {
      "entropy": 5.977344179153443,
      "epoch": 0.2713523131672598,
      "grad_norm": 1.0,
      "learning_rate": 0.000499663383807948,
      "loss": 5.9049,
      "mean_token_accuracy": 0.1500097319483757,
      "num_tokens": 6270582.0,
      "step": 2745
    },
    {
      "entropy": 5.968403148651123,
      "epoch": 0.27184657967576115,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004996614513872781,
      "loss": 5.9016,
      "mean_token_accuracy": 0.15006383657455444,
      "num_tokens": 6281088.0,
      "step": 2750
    },
    {
      "entropy": 5.981626415252686,
      "epoch": 0.2723408461842626,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004996595134399017,
      "loss": 6.0555,
      "mean_token_accuracy": 0.140472724288702,
      "num_tokens": 6292920.0,
      "step": 2755
    },
    {
      "entropy": 6.020872259140015,
      "epoch": 0.27283511269276395,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004996575699658664,
      "loss": 5.9111,
      "mean_token_accuracy": 0.15108120441436768,
      "num_tokens": 6304980.0,
      "step": 2760
    },
    {
      "entropy": 5.918898677825927,
      "epoch": 0.2733293792012653,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004996556209652199,
      "loss": 5.8999,
      "mean_token_accuracy": 0.15111031085252763,
      "num_tokens": 6316175.0,
      "step": 2765
    },
    {
      "entropy": 5.976380968093872,
      "epoch": 0.2738236457097667,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004996536664380104,
      "loss": 5.9517,
      "mean_token_accuracy": 0.15452203899621964,
      "num_tokens": 6327590.0,
      "step": 2770
    },
    {
      "entropy": 5.924459648132324,
      "epoch": 0.27431791221826807,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004996517063842859,
      "loss": 5.9716,
      "mean_token_accuracy": 0.1498926803469658,
      "num_tokens": 6339066.0,
      "step": 2775
    },
    {
      "entropy": 6.0026692867279055,
      "epoch": 0.2748121787267695,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004996497408040944,
      "loss": 5.9024,
      "mean_token_accuracy": 0.15032327100634574,
      "num_tokens": 6349485.0,
      "step": 2780
    },
    {
      "entropy": 5.844460916519165,
      "epoch": 0.27530644523527087,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004996477696974845,
      "loss": 5.8657,
      "mean_token_accuracy": 0.15661628544330597,
      "num_tokens": 6360775.0,
      "step": 2785
    },
    {
      "entropy": 5.9499735832214355,
      "epoch": 0.27580071174377224,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004996457930645046,
      "loss": 5.883,
      "mean_token_accuracy": 0.148551806807518,
      "num_tokens": 6371645.0,
      "step": 2790
    },
    {
      "entropy": 5.961523580551147,
      "epoch": 0.2762949782522736,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004996438109052034,
      "loss": 5.9074,
      "mean_token_accuracy": 0.15085601136088372,
      "num_tokens": 6383877.0,
      "step": 2795
    },
    {
      "entropy": 5.9515691757202145,
      "epoch": 0.276789244760775,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004996418232196296,
      "loss": 5.9828,
      "mean_token_accuracy": 0.14350538998842238,
      "num_tokens": 6394488.0,
      "step": 2800
    },
    {
      "entropy": 5.979497003555298,
      "epoch": 0.2772835112692764,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004996398300078321,
      "loss": 5.9065,
      "mean_token_accuracy": 0.15064874663949013,
      "num_tokens": 6404948.0,
      "step": 2805
    },
    {
      "entropy": 5.974899482727051,
      "epoch": 0.2777777777777778,
      "grad_norm": 1.0,
      "learning_rate": 0.0004996378312698598,
      "loss": 5.9471,
      "mean_token_accuracy": 0.1533238559961319,
      "num_tokens": 6416273.0,
      "step": 2810
    },
    {
      "entropy": 6.013521671295166,
      "epoch": 0.27827204428627916,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004996358270057622,
      "loss": 5.9557,
      "mean_token_accuracy": 0.1466079130768776,
      "num_tokens": 6427091.0,
      "step": 2815
    },
    {
      "entropy": 5.9516332149505615,
      "epoch": 0.27876631079478054,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004996338172155883,
      "loss": 5.9126,
      "mean_token_accuracy": 0.14914426058530808,
      "num_tokens": 6438254.0,
      "step": 2820
    },
    {
      "entropy": 5.89876914024353,
      "epoch": 0.2792605773032819,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004996318018993876,
      "loss": 5.8239,
      "mean_token_accuracy": 0.1530715063214302,
      "num_tokens": 6450618.0,
      "step": 2825
    },
    {
      "entropy": 5.980556011199951,
      "epoch": 0.27975484381178334,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004996297810572098,
      "loss": 5.8956,
      "mean_token_accuracy": 0.1491038627922535,
      "num_tokens": 6462467.0,
      "step": 2830
    },
    {
      "entropy": 5.9045891761779785,
      "epoch": 0.2802491103202847,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004996277546891045,
      "loss": 5.864,
      "mean_token_accuracy": 0.15746333003044127,
      "num_tokens": 6474206.0,
      "step": 2835
    },
    {
      "entropy": 5.942203712463379,
      "epoch": 0.2807433768287861,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004996257227951216,
      "loss": 5.9691,
      "mean_token_accuracy": 0.1443627193570137,
      "num_tokens": 6486405.0,
      "step": 2840
    },
    {
      "entropy": 5.967880725860596,
      "epoch": 0.28123764333728746,
      "grad_norm": 1.09375,
      "learning_rate": 0.000499623685375311,
      "loss": 5.9336,
      "mean_token_accuracy": 0.14780732095241547,
      "num_tokens": 6496949.0,
      "step": 2845
    },
    {
      "entropy": 5.923089170455933,
      "epoch": 0.28173190984578883,
      "grad_norm": 1.0,
      "learning_rate": 0.000499621642429723,
      "loss": 5.874,
      "mean_token_accuracy": 0.1550042062997818,
      "num_tokens": 6508913.0,
      "step": 2850
    },
    {
      "entropy": 5.956121826171875,
      "epoch": 0.28222617635429026,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004996195939584078,
      "loss": 5.9019,
      "mean_token_accuracy": 0.14797470420598985,
      "num_tokens": 6519837.0,
      "step": 2855
    },
    {
      "entropy": 5.941003894805908,
      "epoch": 0.28272044286279163,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004996175399614156,
      "loss": 5.872,
      "mean_token_accuracy": 0.14837601631879807,
      "num_tokens": 6531409.0,
      "step": 2860
    },
    {
      "entropy": 5.937231063842773,
      "epoch": 0.283214709371293,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004996154804387972,
      "loss": 5.9121,
      "mean_token_accuracy": 0.15122439116239547,
      "num_tokens": 6543433.0,
      "step": 2865
    },
    {
      "entropy": 5.91095666885376,
      "epoch": 0.2837089758797944,
      "grad_norm": 1.140625,
      "learning_rate": 0.000499613415390603,
      "loss": 5.9883,
      "mean_token_accuracy": 0.14575250148773194,
      "num_tokens": 6554511.0,
      "step": 2870
    },
    {
      "entropy": 6.029332876205444,
      "epoch": 0.28420324238829575,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000499611344816884,
      "loss": 5.9113,
      "mean_token_accuracy": 0.1500216841697693,
      "num_tokens": 6564591.0,
      "step": 2875
    },
    {
      "entropy": 5.943623399734497,
      "epoch": 0.2846975088967972,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000499609268717691,
      "loss": 5.8826,
      "mean_token_accuracy": 0.15444658771157266,
      "num_tokens": 6575684.0,
      "step": 2880
    },
    {
      "entropy": 6.002884864807129,
      "epoch": 0.28519177540529855,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004996071870930753,
      "loss": 6.0142,
      "mean_token_accuracy": 0.1460142984986305,
      "num_tokens": 6588504.0,
      "step": 2885
    },
    {
      "entropy": 5.843828010559082,
      "epoch": 0.2856860419137999,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004996050999430878,
      "loss": 5.8341,
      "mean_token_accuracy": 0.1514135092496872,
      "num_tokens": 6599969.0,
      "step": 2890
    },
    {
      "entropy": 5.977163600921631,
      "epoch": 0.2861803084223013,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004996030072677802,
      "loss": 5.9088,
      "mean_token_accuracy": 0.15263523682951927,
      "num_tokens": 6611430.0,
      "step": 2895
    },
    {
      "entropy": 5.925257349014283,
      "epoch": 0.2866745749308027,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004996009090672037,
      "loss": 5.8906,
      "mean_token_accuracy": 0.14462630748748778,
      "num_tokens": 6623247.0,
      "step": 2900
    },
    {
      "entropy": 5.96600112915039,
      "epoch": 0.28716884143930405,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004995988053414098,
      "loss": 6.0225,
      "mean_token_accuracy": 0.14478808343410493,
      "num_tokens": 6635508.0,
      "step": 2905
    },
    {
      "entropy": 5.954598236083984,
      "epoch": 0.2876631079478055,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004995966960904506,
      "loss": 5.9278,
      "mean_token_accuracy": 0.14939955547451972,
      "num_tokens": 6646020.0,
      "step": 2910
    },
    {
      "entropy": 5.978266286849975,
      "epoch": 0.28815737445630685,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004995945813143778,
      "loss": 5.9568,
      "mean_token_accuracy": 0.14021606743335724,
      "num_tokens": 6658194.0,
      "step": 2915
    },
    {
      "entropy": 5.96915636062622,
      "epoch": 0.2886516409648082,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004995924610132435,
      "loss": 5.9182,
      "mean_token_accuracy": 0.1475129932165146,
      "num_tokens": 6670582.0,
      "step": 2920
    },
    {
      "entropy": 5.928704261779785,
      "epoch": 0.2891459074733096,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004995903351870998,
      "loss": 5.8759,
      "mean_token_accuracy": 0.14914150834083556,
      "num_tokens": 6680836.0,
      "step": 2925
    },
    {
      "entropy": 6.073338317871094,
      "epoch": 0.28964017398181097,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000499588203835999,
      "loss": 5.9717,
      "mean_token_accuracy": 0.14328644201159477,
      "num_tokens": 6693045.0,
      "step": 2930
    },
    {
      "entropy": 5.933247041702271,
      "epoch": 0.2901344404903124,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004995860669599935,
      "loss": 5.9092,
      "mean_token_accuracy": 0.14393445253372192,
      "num_tokens": 6704006.0,
      "step": 2935
    },
    {
      "entropy": 5.837153053283691,
      "epoch": 0.29062870699881377,
      "grad_norm": 0.96875,
      "learning_rate": 0.000499583924559136,
      "loss": 5.8793,
      "mean_token_accuracy": 0.14840964674949647,
      "num_tokens": 6717027.0,
      "step": 2940
    },
    {
      "entropy": 5.981049108505249,
      "epoch": 0.29112297350731514,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000499581776633479,
      "loss": 5.8749,
      "mean_token_accuracy": 0.15013007670640946,
      "num_tokens": 6728107.0,
      "step": 2945
    },
    {
      "entropy": 5.88144817352295,
      "epoch": 0.2916172400158165,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004995796231830756,
      "loss": 5.8902,
      "mean_token_accuracy": 0.15416626930236815,
      "num_tokens": 6738628.0,
      "step": 2950
    },
    {
      "entropy": 5.934478712081909,
      "epoch": 0.2921115065243179,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004995774642079786,
      "loss": 5.8421,
      "mean_token_accuracy": 0.1537696972489357,
      "num_tokens": 6749416.0,
      "step": 2955
    },
    {
      "entropy": 5.905332374572754,
      "epoch": 0.2926057730328193,
      "grad_norm": 1.0625,
      "learning_rate": 0.000499575299708241,
      "loss": 5.947,
      "mean_token_accuracy": 0.14088274016976357,
      "num_tokens": 6761663.0,
      "step": 2960
    },
    {
      "entropy": 6.020952272415161,
      "epoch": 0.2931000395413207,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004995731296839163,
      "loss": 5.8909,
      "mean_token_accuracy": 0.1538805305957794,
      "num_tokens": 6773557.0,
      "step": 2965
    },
    {
      "entropy": 5.81088809967041,
      "epoch": 0.29359430604982206,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004995709541350578,
      "loss": 5.8199,
      "mean_token_accuracy": 0.1593501538038254,
      "num_tokens": 6782832.0,
      "step": 2970
    },
    {
      "entropy": 5.9672423839569095,
      "epoch": 0.29408857255832344,
      "grad_norm": 1.109375,
      "learning_rate": 0.000499568773061719,
      "loss": 5.9335,
      "mean_token_accuracy": 0.14496245384216308,
      "num_tokens": 6795165.0,
      "step": 2975
    },
    {
      "entropy": 6.023906803131103,
      "epoch": 0.2945828390668248,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004995665864639534,
      "loss": 5.9486,
      "mean_token_accuracy": 0.1483907014131546,
      "num_tokens": 6807054.0,
      "step": 2980
    },
    {
      "entropy": 5.876065492630005,
      "epoch": 0.29507710557532624,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004995643943418149,
      "loss": 5.7332,
      "mean_token_accuracy": 0.15653893128037452,
      "num_tokens": 6817681.0,
      "step": 2985
    },
    {
      "entropy": 5.863463449478149,
      "epoch": 0.2955713720838276,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004995621966953576,
      "loss": 5.9026,
      "mean_token_accuracy": 0.1518360435962677,
      "num_tokens": 6829387.0,
      "step": 2990
    },
    {
      "entropy": 5.936010360717773,
      "epoch": 0.296065638592329,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004995599935246355,
      "loss": 5.9071,
      "mean_token_accuracy": 0.1520347461104393,
      "num_tokens": 6842023.0,
      "step": 2995
    },
    {
      "entropy": 5.92348346710205,
      "epoch": 0.29655990510083036,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004995577848297026,
      "loss": 5.8992,
      "mean_token_accuracy": 0.1477331593632698,
      "num_tokens": 6853189.0,
      "step": 3000
    },
    {
      "epoch": 0.29655990510083036,
      "eval_entropy": 5.794283665472878,
      "eval_loss": 5.955329895019531,
      "eval_mean_token_accuracy": 0.15471349589865732,
      "eval_num_tokens": 6853189.0,
      "eval_runtime": 26.573,
      "eval_samples_per_second": 1223.537,
      "eval_steps_per_second": 152.975,
      "step": 3000
    },
    {
      "entropy": 5.963679313659668,
      "epoch": 0.29705417160933173,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004995555706106134,
      "loss": 5.9504,
      "mean_token_accuracy": 0.1432880811393261,
      "num_tokens": 6864338.0,
      "step": 3005
    },
    {
      "entropy": 5.9105650901794435,
      "epoch": 0.29754843811783316,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004995533508674223,
      "loss": 5.8262,
      "mean_token_accuracy": 0.15388680547475814,
      "num_tokens": 6875409.0,
      "step": 3010
    },
    {
      "entropy": 5.927374696731567,
      "epoch": 0.29804270462633453,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004995511256001839,
      "loss": 5.8561,
      "mean_token_accuracy": 0.14918482303619385,
      "num_tokens": 6885757.0,
      "step": 3015
    },
    {
      "entropy": 5.894947910308838,
      "epoch": 0.2985369711348359,
      "grad_norm": 1.125,
      "learning_rate": 0.000499548894808953,
      "loss": 5.9246,
      "mean_token_accuracy": 0.149386990070343,
      "num_tokens": 6896440.0,
      "step": 3020
    },
    {
      "entropy": 5.982046318054199,
      "epoch": 0.2990312376433373,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004995466584937846,
      "loss": 5.934,
      "mean_token_accuracy": 0.14567233771085739,
      "num_tokens": 6907319.0,
      "step": 3025
    },
    {
      "entropy": 5.953330326080322,
      "epoch": 0.29952550415183865,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004995444166547333,
      "loss": 5.8928,
      "mean_token_accuracy": 0.14992106705904007,
      "num_tokens": 6918683.0,
      "step": 3030
    },
    {
      "entropy": 5.879758739471436,
      "epoch": 0.3000197706603401,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004995421692918547,
      "loss": 5.8745,
      "mean_token_accuracy": 0.15101571381092072,
      "num_tokens": 6929974.0,
      "step": 3035
    },
    {
      "entropy": 5.998798656463623,
      "epoch": 0.30051403716884145,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004995399164052038,
      "loss": 5.8816,
      "mean_token_accuracy": 0.14776666909456254,
      "num_tokens": 6941213.0,
      "step": 3040
    },
    {
      "entropy": 6.010467910766602,
      "epoch": 0.3010083036773428,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004995376579948362,
      "loss": 5.9299,
      "mean_token_accuracy": 0.149107863008976,
      "num_tokens": 6952880.0,
      "step": 3045
    },
    {
      "entropy": 5.940840101242065,
      "epoch": 0.3015025701858442,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004995353940608074,
      "loss": 5.916,
      "mean_token_accuracy": 0.14906010404229164,
      "num_tokens": 6964344.0,
      "step": 3050
    },
    {
      "entropy": 5.867275619506836,
      "epoch": 0.30199683669434557,
      "grad_norm": 1.046875,
      "learning_rate": 0.000499533124603173,
      "loss": 5.8496,
      "mean_token_accuracy": 0.15172645151615144,
      "num_tokens": 6976042.0,
      "step": 3055
    },
    {
      "entropy": 5.927947092056274,
      "epoch": 0.302491103202847,
      "grad_norm": 1.046875,
      "learning_rate": 0.000499530849621989,
      "loss": 5.8254,
      "mean_token_accuracy": 0.1564239665865898,
      "num_tokens": 6988137.0,
      "step": 3060
    },
    {
      "entropy": 5.871209049224854,
      "epoch": 0.3029853697113484,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004995285691173111,
      "loss": 5.8455,
      "mean_token_accuracy": 0.15776384100317956,
      "num_tokens": 6999858.0,
      "step": 3065
    },
    {
      "entropy": 5.956633234024048,
      "epoch": 0.30347963621984975,
      "grad_norm": 1.125,
      "learning_rate": 0.0004995262830891957,
      "loss": 5.9192,
      "mean_token_accuracy": 0.1524971157312393,
      "num_tokens": 7012756.0,
      "step": 3070
    },
    {
      "entropy": 5.922817134857178,
      "epoch": 0.3039739027283511,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004995239915376988,
      "loss": 5.928,
      "mean_token_accuracy": 0.1458661213517189,
      "num_tokens": 7024769.0,
      "step": 3075
    },
    {
      "entropy": 5.925840139389038,
      "epoch": 0.3044681692368525,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004995216944628769,
      "loss": 5.9021,
      "mean_token_accuracy": 0.1427825465798378,
      "num_tokens": 7035422.0,
      "step": 3080
    },
    {
      "entropy": 5.907672977447509,
      "epoch": 0.3049624357453539,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004995193918647865,
      "loss": 5.8496,
      "mean_token_accuracy": 0.1537895068526268,
      "num_tokens": 7046923.0,
      "step": 3085
    },
    {
      "entropy": 6.0074094295501705,
      "epoch": 0.3054567022538553,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004995170837434844,
      "loss": 5.8896,
      "mean_token_accuracy": 0.1455712527036667,
      "num_tokens": 7058241.0,
      "step": 3090
    },
    {
      "entropy": 5.843660068511963,
      "epoch": 0.30595096876235667,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004995147700990269,
      "loss": 5.9194,
      "mean_token_accuracy": 0.15101224333047866,
      "num_tokens": 7070028.0,
      "step": 3095
    },
    {
      "entropy": 5.969692516326904,
      "epoch": 0.30644523527085804,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004995124509314714,
      "loss": 5.898,
      "mean_token_accuracy": 0.14975875839591027,
      "num_tokens": 7082641.0,
      "step": 3100
    },
    {
      "entropy": 5.934259939193725,
      "epoch": 0.3069395017793594,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004995101262408748,
      "loss": 5.8971,
      "mean_token_accuracy": 0.14581625610589982,
      "num_tokens": 7094462.0,
      "step": 3105
    },
    {
      "entropy": 5.879360914230347,
      "epoch": 0.3074337682878608,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004995077960272943,
      "loss": 5.8656,
      "mean_token_accuracy": 0.15483699291944503,
      "num_tokens": 7106598.0,
      "step": 3110
    },
    {
      "entropy": 5.908220291137695,
      "epoch": 0.3079280347963622,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004995054602907871,
      "loss": 5.7884,
      "mean_token_accuracy": 0.15013569295406343,
      "num_tokens": 7117006.0,
      "step": 3115
    },
    {
      "entropy": 5.8303180694580075,
      "epoch": 0.3084223013048636,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004995031190314107,
      "loss": 5.8671,
      "mean_token_accuracy": 0.15000214129686357,
      "num_tokens": 7128071.0,
      "step": 3120
    },
    {
      "entropy": 5.932742738723755,
      "epoch": 0.30891656781336496,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004995007722492228,
      "loss": 5.89,
      "mean_token_accuracy": 0.1478637784719467,
      "num_tokens": 7139896.0,
      "step": 3125
    },
    {
      "entropy": 5.860946559906006,
      "epoch": 0.30941083432186633,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004994984199442811,
      "loss": 5.8406,
      "mean_token_accuracy": 0.1579609915614128,
      "num_tokens": 7151479.0,
      "step": 3130
    },
    {
      "entropy": 5.981082534790039,
      "epoch": 0.3099051008303677,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004994960621166434,
      "loss": 5.8752,
      "mean_token_accuracy": 0.1548340156674385,
      "num_tokens": 7162942.0,
      "step": 3135
    },
    {
      "entropy": 5.838286972045898,
      "epoch": 0.31039936733886914,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004994936987663678,
      "loss": 5.8426,
      "mean_token_accuracy": 0.16279832720756532,
      "num_tokens": 7174572.0,
      "step": 3140
    },
    {
      "entropy": 5.894599485397339,
      "epoch": 0.3108936338473705,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004994913298935124,
      "loss": 5.8984,
      "mean_token_accuracy": 0.14952836856245993,
      "num_tokens": 7186352.0,
      "step": 3145
    },
    {
      "entropy": 5.913238382339477,
      "epoch": 0.3113879003558719,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004994889554981353,
      "loss": 5.8677,
      "mean_token_accuracy": 0.1546166017651558,
      "num_tokens": 7197849.0,
      "step": 3150
    },
    {
      "entropy": 5.900243902206421,
      "epoch": 0.31188216686437326,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004994865755802953,
      "loss": 5.8454,
      "mean_token_accuracy": 0.14976087883114814,
      "num_tokens": 7207897.0,
      "step": 3155
    },
    {
      "entropy": 5.965626001358032,
      "epoch": 0.31237643337287463,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004994841901400505,
      "loss": 5.9604,
      "mean_token_accuracy": 0.14765686094760894,
      "num_tokens": 7219148.0,
      "step": 3160
    },
    {
      "entropy": 6.065278720855713,
      "epoch": 0.31287069988137606,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004994817991774599,
      "loss": 6.0262,
      "mean_token_accuracy": 0.13792147636413574,
      "num_tokens": 7230690.0,
      "step": 3165
    },
    {
      "entropy": 6.003130722045898,
      "epoch": 0.31336496638987743,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004994794026925822,
      "loss": 5.8691,
      "mean_token_accuracy": 0.15180785357952117,
      "num_tokens": 7240935.0,
      "step": 3170
    },
    {
      "entropy": 5.883154249191284,
      "epoch": 0.3138592328983788,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004994770006854763,
      "loss": 5.9264,
      "mean_token_accuracy": 0.15141893699765205,
      "num_tokens": 7253345.0,
      "step": 3175
    },
    {
      "entropy": 6.033541965484619,
      "epoch": 0.3143534994068802,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004994745931562015,
      "loss": 5.9278,
      "mean_token_accuracy": 0.1564277797937393,
      "num_tokens": 7265355.0,
      "step": 3180
    },
    {
      "entropy": 5.799361228942871,
      "epoch": 0.31484776591538155,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004994721801048168,
      "loss": 5.8203,
      "mean_token_accuracy": 0.15881146639585494,
      "num_tokens": 7277355.0,
      "step": 3185
    },
    {
      "entropy": 5.938270950317383,
      "epoch": 0.315342032423883,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004994697615313817,
      "loss": 5.884,
      "mean_token_accuracy": 0.15086214244365692,
      "num_tokens": 7288754.0,
      "step": 3190
    },
    {
      "entropy": 5.974021053314209,
      "epoch": 0.31583629893238435,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004994673374359556,
      "loss": 5.9535,
      "mean_token_accuracy": 0.14664196968078613,
      "num_tokens": 7301158.0,
      "step": 3195
    },
    {
      "entropy": 5.9225036144256595,
      "epoch": 0.3163305654408857,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004994649078185982,
      "loss": 5.9181,
      "mean_token_accuracy": 0.14927734285593033,
      "num_tokens": 7314420.0,
      "step": 3200
    },
    {
      "entropy": 5.934207391738892,
      "epoch": 0.3168248319493871,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004994624726793693,
      "loss": 5.8647,
      "mean_token_accuracy": 0.15222589820623397,
      "num_tokens": 7325459.0,
      "step": 3205
    },
    {
      "entropy": 5.925555992126465,
      "epoch": 0.31731909845788847,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004994600320183286,
      "loss": 5.8421,
      "mean_token_accuracy": 0.15424371659755706,
      "num_tokens": 7336016.0,
      "step": 3210
    },
    {
      "entropy": 5.891087865829467,
      "epoch": 0.3178133649663899,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004994575858355364,
      "loss": 5.8343,
      "mean_token_accuracy": 0.15168708860874175,
      "num_tokens": 7346716.0,
      "step": 3215
    },
    {
      "entropy": 5.9876645565032955,
      "epoch": 0.3183076314748913,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004994551341310526,
      "loss": 5.9459,
      "mean_token_accuracy": 0.1490081638097763,
      "num_tokens": 7357246.0,
      "step": 3220
    },
    {
      "entropy": 5.9510910987854,
      "epoch": 0.31880189798339265,
      "grad_norm": 1.25,
      "learning_rate": 0.0004994526769049379,
      "loss": 5.9666,
      "mean_token_accuracy": 0.1405780427157879,
      "num_tokens": 7369173.0,
      "step": 3225
    },
    {
      "entropy": 5.986663293838501,
      "epoch": 0.319296164491894,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004994502141572524,
      "loss": 5.9451,
      "mean_token_accuracy": 0.13844343274831772,
      "num_tokens": 7381118.0,
      "step": 3230
    },
    {
      "entropy": 5.9236208438873295,
      "epoch": 0.3197904310003954,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004994477458880568,
      "loss": 5.7938,
      "mean_token_accuracy": 0.15767068266868592,
      "num_tokens": 7391874.0,
      "step": 3235
    },
    {
      "entropy": 5.865980625152588,
      "epoch": 0.3202846975088968,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004994452720974119,
      "loss": 5.8872,
      "mean_token_accuracy": 0.15512483119964598,
      "num_tokens": 7403474.0,
      "step": 3240
    },
    {
      "entropy": 5.942899942398071,
      "epoch": 0.3207789640173982,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004994427927853785,
      "loss": 5.7869,
      "mean_token_accuracy": 0.15654317587614058,
      "num_tokens": 7415173.0,
      "step": 3245
    },
    {
      "entropy": 5.843418598175049,
      "epoch": 0.32127323052589957,
      "grad_norm": 1.125,
      "learning_rate": 0.0004994403079520175,
      "loss": 5.8195,
      "mean_token_accuracy": 0.15521105378866196,
      "num_tokens": 7426965.0,
      "step": 3250
    },
    {
      "entropy": 5.930048942565918,
      "epoch": 0.32176749703440094,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004994378175973902,
      "loss": 5.7899,
      "mean_token_accuracy": 0.15678596049547194,
      "num_tokens": 7438162.0,
      "step": 3255
    },
    {
      "entropy": 5.909450387954712,
      "epoch": 0.3222617635429023,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004994353217215577,
      "loss": 5.8716,
      "mean_token_accuracy": 0.1480770692229271,
      "num_tokens": 7450241.0,
      "step": 3260
    },
    {
      "entropy": 5.9811969757080075,
      "epoch": 0.32275603005140374,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004994328203245814,
      "loss": 5.9739,
      "mean_token_accuracy": 0.1421304538846016,
      "num_tokens": 7462236.0,
      "step": 3265
    },
    {
      "entropy": 5.981936073303222,
      "epoch": 0.3232502965599051,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004994303134065231,
      "loss": 5.8787,
      "mean_token_accuracy": 0.14774280935525894,
      "num_tokens": 7472811.0,
      "step": 3270
    },
    {
      "entropy": 5.823878812789917,
      "epoch": 0.3237445630684065,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004994278009674442,
      "loss": 5.7781,
      "mean_token_accuracy": 0.16131176203489303,
      "num_tokens": 7484837.0,
      "step": 3275
    },
    {
      "entropy": 5.996675300598144,
      "epoch": 0.32423882957690786,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004994252830074064,
      "loss": 5.8776,
      "mean_token_accuracy": 0.14760498106479644,
      "num_tokens": 7496712.0,
      "step": 3280
    },
    {
      "entropy": 5.824530220031738,
      "epoch": 0.32473309608540923,
      "grad_norm": 1.078125,
      "learning_rate": 0.000499422759526472,
      "loss": 5.8799,
      "mean_token_accuracy": 0.1516605421900749,
      "num_tokens": 7508014.0,
      "step": 3285
    },
    {
      "entropy": 5.916428327560425,
      "epoch": 0.32522736259391066,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004994202305247029,
      "loss": 5.9297,
      "mean_token_accuracy": 0.14520037919282913,
      "num_tokens": 7519466.0,
      "step": 3290
    },
    {
      "entropy": 6.085912036895752,
      "epoch": 0.32572162910241204,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004994176960021613,
      "loss": 5.9715,
      "mean_token_accuracy": 0.1455385245382786,
      "num_tokens": 7531661.0,
      "step": 3295
    },
    {
      "entropy": 5.927366018295288,
      "epoch": 0.3262158956109134,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004994151559589095,
      "loss": 5.8879,
      "mean_token_accuracy": 0.14917458668351175,
      "num_tokens": 7544096.0,
      "step": 3300
    },
    {
      "entropy": 5.924072742462158,
      "epoch": 0.3267101621194148,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004994126103950101,
      "loss": 5.863,
      "mean_token_accuracy": 0.14948750287294388,
      "num_tokens": 7556022.0,
      "step": 3305
    },
    {
      "entropy": 5.904584789276123,
      "epoch": 0.32720442862791616,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004994100593105257,
      "loss": 5.835,
      "mean_token_accuracy": 0.15778674483299254,
      "num_tokens": 7567399.0,
      "step": 3310
    },
    {
      "entropy": 5.845813608169555,
      "epoch": 0.3276986951364176,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004994075027055189,
      "loss": 5.9029,
      "mean_token_accuracy": 0.150156831741333,
      "num_tokens": 7580673.0,
      "step": 3315
    },
    {
      "entropy": 5.995374393463135,
      "epoch": 0.32819296164491896,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004994049405800529,
      "loss": 5.849,
      "mean_token_accuracy": 0.14884991571307182,
      "num_tokens": 7591799.0,
      "step": 3320
    },
    {
      "entropy": 5.906673097610474,
      "epoch": 0.32868722815342033,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004994023729341904,
      "loss": 5.8885,
      "mean_token_accuracy": 0.1490943245589733,
      "num_tokens": 7603663.0,
      "step": 3325
    },
    {
      "entropy": 5.905765390396118,
      "epoch": 0.3291814946619217,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004993997997679949,
      "loss": 5.8288,
      "mean_token_accuracy": 0.15654058754444122,
      "num_tokens": 7615651.0,
      "step": 3330
    },
    {
      "entropy": 5.823675203323364,
      "epoch": 0.3296757611704231,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004993972210815292,
      "loss": 5.8597,
      "mean_token_accuracy": 0.15760498493909836,
      "num_tokens": 7628033.0,
      "step": 3335
    },
    {
      "entropy": 5.996159982681275,
      "epoch": 0.33017002767892445,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004993946368748573,
      "loss": 5.9642,
      "mean_token_accuracy": 0.14904893189668655,
      "num_tokens": 7640227.0,
      "step": 3340
    },
    {
      "entropy": 5.973145627975464,
      "epoch": 0.3306642941874259,
      "grad_norm": 1.125,
      "learning_rate": 0.0004993920471480424,
      "loss": 5.8588,
      "mean_token_accuracy": 0.155265648663044,
      "num_tokens": 7651684.0,
      "step": 3345
    },
    {
      "entropy": 5.897155094146728,
      "epoch": 0.33115856069592725,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004993894519011484,
      "loss": 5.8992,
      "mean_token_accuracy": 0.15093035399913787,
      "num_tokens": 7662721.0,
      "step": 3350
    },
    {
      "entropy": 5.908133935928345,
      "epoch": 0.3316528272044286,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004993868511342391,
      "loss": 5.8487,
      "mean_token_accuracy": 0.15349284261465074,
      "num_tokens": 7674060.0,
      "step": 3355
    },
    {
      "entropy": 5.858733654022217,
      "epoch": 0.33214709371293,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004993842448473783,
      "loss": 5.8045,
      "mean_token_accuracy": 0.15487059578299522,
      "num_tokens": 7685054.0,
      "step": 3360
    },
    {
      "entropy": 5.886049079895019,
      "epoch": 0.33264136022143137,
      "grad_norm": 1.25,
      "learning_rate": 0.0004993816330406304,
      "loss": 5.9195,
      "mean_token_accuracy": 0.14408770725131034,
      "num_tokens": 7696554.0,
      "step": 3365
    },
    {
      "entropy": 6.02078652381897,
      "epoch": 0.3331356267299328,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004993790157140595,
      "loss": 5.7918,
      "mean_token_accuracy": 0.15052541792392732,
      "num_tokens": 7707000.0,
      "step": 3370
    },
    {
      "entropy": 5.9085273265838625,
      "epoch": 0.33362989323843417,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004993763928677299,
      "loss": 5.8716,
      "mean_token_accuracy": 0.15730493217706681,
      "num_tokens": 7718898.0,
      "step": 3375
    },
    {
      "entropy": 5.933470582962036,
      "epoch": 0.33412415974693555,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004993737645017064,
      "loss": 5.9046,
      "mean_token_accuracy": 0.15121666342020035,
      "num_tokens": 7730493.0,
      "step": 3380
    },
    {
      "entropy": 5.896996593475341,
      "epoch": 0.3346184262554369,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004993711306160534,
      "loss": 5.8152,
      "mean_token_accuracy": 0.15182201862335204,
      "num_tokens": 7742177.0,
      "step": 3385
    },
    {
      "entropy": 5.885739374160766,
      "epoch": 0.3351126927639383,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004993684912108358,
      "loss": 5.8901,
      "mean_token_accuracy": 0.1533852696418762,
      "num_tokens": 7753386.0,
      "step": 3390
    },
    {
      "entropy": 5.854862308502197,
      "epoch": 0.3356069592724397,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004993658462861184,
      "loss": 5.7926,
      "mean_token_accuracy": 0.15316681638360025,
      "num_tokens": 7764259.0,
      "step": 3395
    },
    {
      "entropy": 5.915011262893676,
      "epoch": 0.3361012257809411,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004993631958419666,
      "loss": 5.7892,
      "mean_token_accuracy": 0.15640075802803038,
      "num_tokens": 7775853.0,
      "step": 3400
    },
    {
      "entropy": 5.902416372299195,
      "epoch": 0.33659549228944247,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004993605398784451,
      "loss": 5.9146,
      "mean_token_accuracy": 0.139879260212183,
      "num_tokens": 7787116.0,
      "step": 3405
    },
    {
      "entropy": 5.956493091583252,
      "epoch": 0.33708975879794384,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004993578783956198,
      "loss": 5.9393,
      "mean_token_accuracy": 0.1419096127152443,
      "num_tokens": 7799666.0,
      "step": 3410
    },
    {
      "entropy": 5.90988245010376,
      "epoch": 0.3375840253064452,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004993552113935557,
      "loss": 5.8263,
      "mean_token_accuracy": 0.1583866521716118,
      "num_tokens": 7812454.0,
      "step": 3415
    },
    {
      "entropy": 5.916579532623291,
      "epoch": 0.33807829181494664,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004993525388723186,
      "loss": 5.8834,
      "mean_token_accuracy": 0.1530376985669136,
      "num_tokens": 7824427.0,
      "step": 3420
    },
    {
      "entropy": 5.901572895050049,
      "epoch": 0.338572558323448,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004993498608319742,
      "loss": 5.8844,
      "mean_token_accuracy": 0.15276393666863441,
      "num_tokens": 7835382.0,
      "step": 3425
    },
    {
      "entropy": 5.859151458740234,
      "epoch": 0.3390668248319494,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004993471772725886,
      "loss": 5.7989,
      "mean_token_accuracy": 0.15550214797258377,
      "num_tokens": 7846618.0,
      "step": 3430
    },
    {
      "entropy": 5.901988458633423,
      "epoch": 0.33956109134045076,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004993444881942274,
      "loss": 5.8146,
      "mean_token_accuracy": 0.15682211965322496,
      "num_tokens": 7859160.0,
      "step": 3435
    },
    {
      "entropy": 5.8803332328796385,
      "epoch": 0.34005535784895213,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004993417935969572,
      "loss": 5.8434,
      "mean_token_accuracy": 0.15452905595302582,
      "num_tokens": 7870628.0,
      "step": 3440
    },
    {
      "entropy": 5.94140362739563,
      "epoch": 0.34054962435745356,
      "grad_norm": 1.109375,
      "learning_rate": 0.000499339093480844,
      "loss": 5.8743,
      "mean_token_accuracy": 0.1545655608177185,
      "num_tokens": 7883016.0,
      "step": 3445
    },
    {
      "entropy": 5.867698383331299,
      "epoch": 0.34104389086595494,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004993363878459542,
      "loss": 5.8751,
      "mean_token_accuracy": 0.14724137187004088,
      "num_tokens": 7895940.0,
      "step": 3450
    },
    {
      "entropy": 5.9430724620819095,
      "epoch": 0.3415381573744563,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004993336766923546,
      "loss": 5.9059,
      "mean_token_accuracy": 0.14910922944545746,
      "num_tokens": 7907920.0,
      "step": 3455
    },
    {
      "entropy": 5.952860927581787,
      "epoch": 0.3420324238829577,
      "grad_norm": 1.375,
      "learning_rate": 0.0004993309600201117,
      "loss": 5.889,
      "mean_token_accuracy": 0.15425898805260657,
      "num_tokens": 7920181.0,
      "step": 3460
    },
    {
      "entropy": 5.888333415985107,
      "epoch": 0.34252669039145905,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004993282378292923,
      "loss": 5.8396,
      "mean_token_accuracy": 0.1592525601387024,
      "num_tokens": 7931135.0,
      "step": 3465
    },
    {
      "entropy": 5.885323476791382,
      "epoch": 0.3430209568999605,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004993255101199635,
      "loss": 5.8211,
      "mean_token_accuracy": 0.15970962271094322,
      "num_tokens": 7941002.0,
      "step": 3470
    },
    {
      "entropy": 5.942181777954102,
      "epoch": 0.34351522340846186,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004993227768921924,
      "loss": 5.792,
      "mean_token_accuracy": 0.1554045557975769,
      "num_tokens": 7950393.0,
      "step": 3475
    },
    {
      "entropy": 5.9116617202758786,
      "epoch": 0.34400948991696323,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004993200381460461,
      "loss": 5.8258,
      "mean_token_accuracy": 0.15245998203754424,
      "num_tokens": 7961468.0,
      "step": 3480
    },
    {
      "entropy": 5.817622375488281,
      "epoch": 0.3445037564254646,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004993172938815922,
      "loss": 5.6706,
      "mean_token_accuracy": 0.1642368882894516,
      "num_tokens": 7973783.0,
      "step": 3485
    },
    {
      "entropy": 5.789644384384156,
      "epoch": 0.344998022933966,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004993145440988979,
      "loss": 5.7316,
      "mean_token_accuracy": 0.16548056602478028,
      "num_tokens": 7984099.0,
      "step": 3490
    },
    {
      "entropy": 5.954108858108521,
      "epoch": 0.3454922894424674,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004993117887980311,
      "loss": 5.8921,
      "mean_token_accuracy": 0.15117180049419404,
      "num_tokens": 7995115.0,
      "step": 3495
    },
    {
      "entropy": 5.920248556137085,
      "epoch": 0.3459865559509688,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004993090279790596,
      "loss": 5.8406,
      "mean_token_accuracy": 0.15169650912284852,
      "num_tokens": 8006103.0,
      "step": 3500
    },
    {
      "entropy": 5.7916463851928714,
      "epoch": 0.34648082245947015,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000499306261642051,
      "loss": 5.8067,
      "mean_token_accuracy": 0.14806041717529297,
      "num_tokens": 8018228.0,
      "step": 3505
    },
    {
      "entropy": 5.884153890609741,
      "epoch": 0.3469750889679715,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004993034897870737,
      "loss": 5.7563,
      "mean_token_accuracy": 0.15497301369905472,
      "num_tokens": 8029808.0,
      "step": 3510
    },
    {
      "entropy": 5.843921661376953,
      "epoch": 0.3474693554764729,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004993007124141958,
      "loss": 5.8536,
      "mean_token_accuracy": 0.14964404553174973,
      "num_tokens": 8041136.0,
      "step": 3515
    },
    {
      "entropy": 5.920845794677734,
      "epoch": 0.3479636219849743,
      "grad_norm": 1.25,
      "learning_rate": 0.0004992979295234854,
      "loss": 5.9013,
      "mean_token_accuracy": 0.14759098887443542,
      "num_tokens": 8053013.0,
      "step": 3520
    },
    {
      "entropy": 5.940861701965332,
      "epoch": 0.3484578884934757,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004992951411150112,
      "loss": 5.8713,
      "mean_token_accuracy": 0.14649871438741685,
      "num_tokens": 8065139.0,
      "step": 3525
    },
    {
      "entropy": 5.860837078094482,
      "epoch": 0.34895215500197707,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004992923471888418,
      "loss": 5.7316,
      "mean_token_accuracy": 0.16246522814035416,
      "num_tokens": 8075719.0,
      "step": 3530
    },
    {
      "entropy": 5.876574087142944,
      "epoch": 0.34944642151047844,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004992895477450459,
      "loss": 5.8621,
      "mean_token_accuracy": 0.15280634313821792,
      "num_tokens": 8087821.0,
      "step": 3535
    },
    {
      "entropy": 5.838806915283203,
      "epoch": 0.3499406880189798,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004992867427836923,
      "loss": 5.755,
      "mean_token_accuracy": 0.15870047360658646,
      "num_tokens": 8099861.0,
      "step": 3540
    },
    {
      "entropy": 6.010531711578369,
      "epoch": 0.3504349545274812,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004992839323048501,
      "loss": 5.9882,
      "mean_token_accuracy": 0.14447013214230536,
      "num_tokens": 8111671.0,
      "step": 3545
    },
    {
      "entropy": 6.0363317966461185,
      "epoch": 0.3509292210359826,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004992811163085882,
      "loss": 5.9404,
      "mean_token_accuracy": 0.150308321416378,
      "num_tokens": 8123569.0,
      "step": 3550
    },
    {
      "entropy": 5.9858965396881105,
      "epoch": 0.351423487544484,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004992782947949761,
      "loss": 5.9541,
      "mean_token_accuracy": 0.15137761533260347,
      "num_tokens": 8135621.0,
      "step": 3555
    },
    {
      "entropy": 5.8561443328857425,
      "epoch": 0.35191775405298537,
      "grad_norm": 1.25,
      "learning_rate": 0.0004992754677640831,
      "loss": 5.7978,
      "mean_token_accuracy": 0.15685692727565764,
      "num_tokens": 8147012.0,
      "step": 3560
    },
    {
      "entropy": 5.908249092102051,
      "epoch": 0.35241202056148674,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004992726352159789,
      "loss": 5.7851,
      "mean_token_accuracy": 0.15655775964260102,
      "num_tokens": 8158686.0,
      "step": 3565
    },
    {
      "entropy": 5.85140061378479,
      "epoch": 0.3529062870699881,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004992697971507331,
      "loss": 5.7901,
      "mean_token_accuracy": 0.15729555934667588,
      "num_tokens": 8170188.0,
      "step": 3570
    },
    {
      "entropy": 5.890997982025146,
      "epoch": 0.35340055357848954,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004992669535684155,
      "loss": 5.8717,
      "mean_token_accuracy": 0.15248715728521348,
      "num_tokens": 8182467.0,
      "step": 3575
    },
    {
      "entropy": 5.957807207107544,
      "epoch": 0.3538948200869909,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004992641044690959,
      "loss": 5.9638,
      "mean_token_accuracy": 0.15112252160906792,
      "num_tokens": 8193433.0,
      "step": 3580
    },
    {
      "entropy": 5.893219804763794,
      "epoch": 0.3543890865954923,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004992612498528447,
      "loss": 5.7478,
      "mean_token_accuracy": 0.1555414766073227,
      "num_tokens": 8204220.0,
      "step": 3585
    },
    {
      "entropy": 5.810119009017944,
      "epoch": 0.35488335310399366,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004992583897197319,
      "loss": 5.7642,
      "mean_token_accuracy": 0.16225289255380632,
      "num_tokens": 8214203.0,
      "step": 3590
    },
    {
      "entropy": 5.9189142227172855,
      "epoch": 0.35537761961249503,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004992555240698279,
      "loss": 5.9038,
      "mean_token_accuracy": 0.14603765159845353,
      "num_tokens": 8225143.0,
      "step": 3595
    },
    {
      "entropy": 5.91175103187561,
      "epoch": 0.35587188612099646,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004992526529032032,
      "loss": 5.7585,
      "mean_token_accuracy": 0.15905050486326217,
      "num_tokens": 8237715.0,
      "step": 3600
    },
    {
      "entropy": 5.805362415313721,
      "epoch": 0.35636615262949783,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004992497762199285,
      "loss": 5.8196,
      "mean_token_accuracy": 0.15873343497514725,
      "num_tokens": 8249645.0,
      "step": 3605
    },
    {
      "entropy": 5.945392084121704,
      "epoch": 0.3568604191379992,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004992468940200745,
      "loss": 5.7897,
      "mean_token_accuracy": 0.1611865848302841,
      "num_tokens": 8260624.0,
      "step": 3610
    },
    {
      "entropy": 5.910581159591675,
      "epoch": 0.3573546856465006,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004992440063037121,
      "loss": 5.8894,
      "mean_token_accuracy": 0.1492436170578003,
      "num_tokens": 8272526.0,
      "step": 3615
    },
    {
      "entropy": 5.777053737640381,
      "epoch": 0.35784895215500195,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004992411130709124,
      "loss": 5.7908,
      "mean_token_accuracy": 0.15661219209432603,
      "num_tokens": 8284270.0,
      "step": 3620
    },
    {
      "entropy": 5.969174575805664,
      "epoch": 0.3583432186635034,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004992382143217466,
      "loss": 5.9021,
      "mean_token_accuracy": 0.15277914106845855,
      "num_tokens": 8296407.0,
      "step": 3625
    },
    {
      "entropy": 5.841032934188843,
      "epoch": 0.35883748517200476,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004992353100562858,
      "loss": 5.805,
      "mean_token_accuracy": 0.16218566447496413,
      "num_tokens": 8307695.0,
      "step": 3630
    },
    {
      "entropy": 5.794250535964966,
      "epoch": 0.35933175168050613,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004992324002746016,
      "loss": 5.7942,
      "mean_token_accuracy": 0.15522915124893188,
      "num_tokens": 8319458.0,
      "step": 3635
    },
    {
      "entropy": 5.872878932952881,
      "epoch": 0.3598260181890075,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004992294849767656,
      "loss": 5.7111,
      "mean_token_accuracy": 0.16143833845853806,
      "num_tokens": 8330284.0,
      "step": 3640
    },
    {
      "entropy": 5.853314065933228,
      "epoch": 0.3603202846975089,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004992265641628495,
      "loss": 5.827,
      "mean_token_accuracy": 0.15445810109376906,
      "num_tokens": 8343622.0,
      "step": 3645
    },
    {
      "entropy": 5.894721937179566,
      "epoch": 0.3608145512060103,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004992236378329252,
      "loss": 5.7864,
      "mean_token_accuracy": 0.1555868223309517,
      "num_tokens": 8354079.0,
      "step": 3650
    },
    {
      "entropy": 5.8952436447143555,
      "epoch": 0.3613088177145117,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004992207059870645,
      "loss": 5.83,
      "mean_token_accuracy": 0.15822544246912001,
      "num_tokens": 8364876.0,
      "step": 3655
    },
    {
      "entropy": 5.934147882461548,
      "epoch": 0.36180308422301305,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004992177686253396,
      "loss": 5.7813,
      "mean_token_accuracy": 0.15549880415201187,
      "num_tokens": 8375095.0,
      "step": 3660
    },
    {
      "entropy": 5.7936859130859375,
      "epoch": 0.3622973507315144,
      "grad_norm": 1.15625,
      "learning_rate": 0.000499214825747823,
      "loss": 5.7544,
      "mean_token_accuracy": 0.15899945050477982,
      "num_tokens": 8385934.0,
      "step": 3665
    },
    {
      "entropy": 5.8251642227172855,
      "epoch": 0.3627916172400158,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004992118773545868,
      "loss": 5.7247,
      "mean_token_accuracy": 0.1615453451871872,
      "num_tokens": 8397018.0,
      "step": 3670
    },
    {
      "entropy": 5.841867876052857,
      "epoch": 0.3632858837485172,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004992089234457036,
      "loss": 5.7961,
      "mean_token_accuracy": 0.1640404224395752,
      "num_tokens": 8406668.0,
      "step": 3675
    },
    {
      "entropy": 5.953409671783447,
      "epoch": 0.3637801502570186,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004992059640212461,
      "loss": 5.8298,
      "mean_token_accuracy": 0.15458669289946556,
      "num_tokens": 8417471.0,
      "step": 3680
    },
    {
      "entropy": 5.856666088104248,
      "epoch": 0.36427441676551997,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004992029990812872,
      "loss": 5.9011,
      "mean_token_accuracy": 0.14456111043691636,
      "num_tokens": 8430073.0,
      "step": 3685
    },
    {
      "entropy": 5.893247652053833,
      "epoch": 0.36476868327402134,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004992000286258997,
      "loss": 5.8086,
      "mean_token_accuracy": 0.15492908284068108,
      "num_tokens": 8442573.0,
      "step": 3690
    },
    {
      "entropy": 5.940594816207886,
      "epoch": 0.3652629497825227,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004991970526551567,
      "loss": 5.8576,
      "mean_token_accuracy": 0.15471384078264236,
      "num_tokens": 8453920.0,
      "step": 3695
    },
    {
      "entropy": 5.851522207260132,
      "epoch": 0.36575721629102415,
      "grad_norm": 1.125,
      "learning_rate": 0.0004991940711691314,
      "loss": 5.8945,
      "mean_token_accuracy": 0.15315308719873427,
      "num_tokens": 8466652.0,
      "step": 3700
    },
    {
      "entropy": 5.93485655784607,
      "epoch": 0.3662514827995255,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004991910841678972,
      "loss": 5.9105,
      "mean_token_accuracy": 0.14977106899023057,
      "num_tokens": 8480203.0,
      "step": 3705
    },
    {
      "entropy": 5.864224863052368,
      "epoch": 0.3667457493080269,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004991880916515275,
      "loss": 5.8174,
      "mean_token_accuracy": 0.15368005335330964,
      "num_tokens": 8491396.0,
      "step": 3710
    },
    {
      "entropy": 5.918771553039551,
      "epoch": 0.36724001581652826,
      "grad_norm": 1.15625,
      "learning_rate": 0.000499185093620096,
      "loss": 5.7838,
      "mean_token_accuracy": 0.15006800889968872,
      "num_tokens": 8501838.0,
      "step": 3715
    },
    {
      "entropy": 5.798188638687134,
      "epoch": 0.36773428232502964,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004991820900736765,
      "loss": 5.6723,
      "mean_token_accuracy": 0.1582162246108055,
      "num_tokens": 8512113.0,
      "step": 3720
    },
    {
      "entropy": 5.831536865234375,
      "epoch": 0.36822854883353107,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004991790810123427,
      "loss": 5.7507,
      "mean_token_accuracy": 0.15408645272254945,
      "num_tokens": 8523888.0,
      "step": 3725
    },
    {
      "entropy": 5.79871392250061,
      "epoch": 0.36872281534203244,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004991760664361688,
      "loss": 5.6859,
      "mean_token_accuracy": 0.1632385030388832,
      "num_tokens": 8536078.0,
      "step": 3730
    },
    {
      "entropy": 5.837257719039917,
      "epoch": 0.3692170818505338,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004991730463452288,
      "loss": 5.8265,
      "mean_token_accuracy": 0.15056723803281785,
      "num_tokens": 8547131.0,
      "step": 3735
    },
    {
      "entropy": 5.847607707977295,
      "epoch": 0.3697113483590352,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004991700207395971,
      "loss": 5.6923,
      "mean_token_accuracy": 0.16566429436206817,
      "num_tokens": 8557269.0,
      "step": 3740
    },
    {
      "entropy": 5.921040964126587,
      "epoch": 0.37020561486753656,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004991669896193482,
      "loss": 5.8815,
      "mean_token_accuracy": 0.146759133040905,
      "num_tokens": 8569193.0,
      "step": 3745
    },
    {
      "entropy": 5.870231437683105,
      "epoch": 0.370699881376038,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004991639529845565,
      "loss": 5.7533,
      "mean_token_accuracy": 0.15931029468774796,
      "num_tokens": 8580888.0,
      "step": 3750
    },
    {
      "entropy": 5.840953636169433,
      "epoch": 0.37119414788453936,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004991609108352968,
      "loss": 5.836,
      "mean_token_accuracy": 0.1561110332608223,
      "num_tokens": 8591868.0,
      "step": 3755
    },
    {
      "entropy": 5.83723406791687,
      "epoch": 0.37168841439304073,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004991578631716438,
      "loss": 5.7962,
      "mean_token_accuracy": 0.15383539870381355,
      "num_tokens": 8602338.0,
      "step": 3760
    },
    {
      "entropy": 5.8507262706756595,
      "epoch": 0.3721826809015421,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004991548099936727,
      "loss": 5.7276,
      "mean_token_accuracy": 0.15699041187763213,
      "num_tokens": 8611698.0,
      "step": 3765
    },
    {
      "entropy": 5.836185693740845,
      "epoch": 0.3726769474100435,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004991517513014585,
      "loss": 5.7327,
      "mean_token_accuracy": 0.1610111728310585,
      "num_tokens": 8623537.0,
      "step": 3770
    },
    {
      "entropy": 5.787095975875855,
      "epoch": 0.37317121391854485,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004991486870950765,
      "loss": 5.7996,
      "mean_token_accuracy": 0.1559009499847889,
      "num_tokens": 8634678.0,
      "step": 3775
    },
    {
      "entropy": 5.864709520339966,
      "epoch": 0.3736654804270463,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004991456173746021,
      "loss": 5.7777,
      "mean_token_accuracy": 0.16089840829372407,
      "num_tokens": 8646207.0,
      "step": 3780
    },
    {
      "entropy": 5.955631351470947,
      "epoch": 0.37415974693554765,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004991425421401105,
      "loss": 5.839,
      "mean_token_accuracy": 0.15324022620916367,
      "num_tokens": 8657069.0,
      "step": 3785
    },
    {
      "entropy": 5.766555738449097,
      "epoch": 0.374654013444049,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004991394613916777,
      "loss": 5.6965,
      "mean_token_accuracy": 0.16205490231513978,
      "num_tokens": 8668296.0,
      "step": 3790
    },
    {
      "entropy": 5.816662788391113,
      "epoch": 0.3751482799525504,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004991363751293795,
      "loss": 5.8514,
      "mean_token_accuracy": 0.15196674913167954,
      "num_tokens": 8680626.0,
      "step": 3795
    },
    {
      "entropy": 5.868446969985962,
      "epoch": 0.3756425464610518,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004991332833532915,
      "loss": 5.7609,
      "mean_token_accuracy": 0.16015817373991012,
      "num_tokens": 8691571.0,
      "step": 3800
    },
    {
      "entropy": 5.863410568237304,
      "epoch": 0.3761368129695532,
      "grad_norm": 1.140625,
      "learning_rate": 0.00049913018606349,
      "loss": 5.814,
      "mean_token_accuracy": 0.16047453731298447,
      "num_tokens": 8703262.0,
      "step": 3805
    },
    {
      "entropy": 5.922005653381348,
      "epoch": 0.3766310794780546,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004991270832600512,
      "loss": 5.9128,
      "mean_token_accuracy": 0.1440843015909195,
      "num_tokens": 8714799.0,
      "step": 3810
    },
    {
      "entropy": 5.948727369308472,
      "epoch": 0.37712534598655595,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004991239749430514,
      "loss": 5.8848,
      "mean_token_accuracy": 0.14975858181715013,
      "num_tokens": 8726417.0,
      "step": 3815
    },
    {
      "entropy": 5.77113709449768,
      "epoch": 0.3776196124950573,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000499120861112567,
      "loss": 5.7504,
      "mean_token_accuracy": 0.16276681572198867,
      "num_tokens": 8738572.0,
      "step": 3820
    },
    {
      "entropy": 5.987140226364136,
      "epoch": 0.3781138790035587,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004991177417686747,
      "loss": 5.9179,
      "mean_token_accuracy": 0.1479107543826103,
      "num_tokens": 8750557.0,
      "step": 3825
    },
    {
      "entropy": 5.889439392089844,
      "epoch": 0.3786081455120601,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004991146169114511,
      "loss": 5.7055,
      "mean_token_accuracy": 0.1570228233933449,
      "num_tokens": 8761921.0,
      "step": 3830
    },
    {
      "entropy": 5.750439548492432,
      "epoch": 0.3791024120205615,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004991114865409732,
      "loss": 5.7991,
      "mean_token_accuracy": 0.15767900198698043,
      "num_tokens": 8772705.0,
      "step": 3835
    },
    {
      "entropy": 5.842856407165527,
      "epoch": 0.37959667852906287,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000499108350657318,
      "loss": 5.8153,
      "mean_token_accuracy": 0.1552213490009308,
      "num_tokens": 8783563.0,
      "step": 3840
    },
    {
      "entropy": 5.858430480957031,
      "epoch": 0.38009094503756424,
      "grad_norm": 1.25,
      "learning_rate": 0.0004991052092605627,
      "loss": 5.7152,
      "mean_token_accuracy": 0.16301678866147995,
      "num_tokens": 8794171.0,
      "step": 3845
    },
    {
      "entropy": 5.80806884765625,
      "epoch": 0.3805852115460656,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004991020623507842,
      "loss": 5.8158,
      "mean_token_accuracy": 0.1521042175590992,
      "num_tokens": 8806487.0,
      "step": 3850
    },
    {
      "entropy": 5.895160150527954,
      "epoch": 0.38107947805456704,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004990989099280604,
      "loss": 5.8211,
      "mean_token_accuracy": 0.1585649460554123,
      "num_tokens": 8817819.0,
      "step": 3855
    },
    {
      "entropy": 5.874980354309082,
      "epoch": 0.3815737445630684,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004990957519924686,
      "loss": 5.764,
      "mean_token_accuracy": 0.15150784254074096,
      "num_tokens": 8828485.0,
      "step": 3860
    },
    {
      "entropy": 5.821585559844971,
      "epoch": 0.3820680110715698,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004990925885440865,
      "loss": 5.8205,
      "mean_token_accuracy": 0.15444822385907173,
      "num_tokens": 8840179.0,
      "step": 3865
    },
    {
      "entropy": 5.953197860717774,
      "epoch": 0.38256227758007116,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004990894195829921,
      "loss": 5.8494,
      "mean_token_accuracy": 0.1579738438129425,
      "num_tokens": 8851813.0,
      "step": 3870
    },
    {
      "entropy": 5.844118452072143,
      "epoch": 0.38305654408857254,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004990862451092631,
      "loss": 5.7774,
      "mean_token_accuracy": 0.15051404684782027,
      "num_tokens": 8863072.0,
      "step": 3875
    },
    {
      "entropy": 5.880052471160889,
      "epoch": 0.38355081059707397,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004990830651229776,
      "loss": 5.7813,
      "mean_token_accuracy": 0.15792119652032852,
      "num_tokens": 8873510.0,
      "step": 3880
    },
    {
      "entropy": 5.8031305313110355,
      "epoch": 0.38404507710557534,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004990798796242142,
      "loss": 5.7356,
      "mean_token_accuracy": 0.15857678204774855,
      "num_tokens": 8884296.0,
      "step": 3885
    },
    {
      "entropy": 5.851049709320068,
      "epoch": 0.3845393436140767,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004990766886130508,
      "loss": 5.7761,
      "mean_token_accuracy": 0.16058651506900787,
      "num_tokens": 8894208.0,
      "step": 3890
    },
    {
      "entropy": 5.988053703308106,
      "epoch": 0.3850336101225781,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004990734920895662,
      "loss": 5.9627,
      "mean_token_accuracy": 0.14808114171028136,
      "num_tokens": 8905703.0,
      "step": 3895
    },
    {
      "entropy": 5.8592668056488035,
      "epoch": 0.38552787663107946,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004990702900538388,
      "loss": 5.7484,
      "mean_token_accuracy": 0.15851551443338394,
      "num_tokens": 8918077.0,
      "step": 3900
    },
    {
      "entropy": 5.788652181625366,
      "epoch": 0.3860221431395809,
      "grad_norm": 1.125,
      "learning_rate": 0.0004990670825059477,
      "loss": 5.7267,
      "mean_token_accuracy": 0.15688996613025666,
      "num_tokens": 8930240.0,
      "step": 3905
    },
    {
      "entropy": 5.900861597061157,
      "epoch": 0.38651640964808226,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004990638694459715,
      "loss": 5.7986,
      "mean_token_accuracy": 0.15847810059785844,
      "num_tokens": 8941421.0,
      "step": 3910
    },
    {
      "entropy": 5.872343826293945,
      "epoch": 0.38701067615658363,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004990606508739893,
      "loss": 5.8115,
      "mean_token_accuracy": 0.15537682622671128,
      "num_tokens": 8952296.0,
      "step": 3915
    },
    {
      "entropy": 5.869143915176392,
      "epoch": 0.387504942665085,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004990574267900804,
      "loss": 5.8186,
      "mean_token_accuracy": 0.15669666230678558,
      "num_tokens": 8963167.0,
      "step": 3920
    },
    {
      "entropy": 5.86326756477356,
      "epoch": 0.3879992091735864,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004990541971943241,
      "loss": 5.787,
      "mean_token_accuracy": 0.15738223493099213,
      "num_tokens": 8974405.0,
      "step": 3925
    },
    {
      "entropy": 5.871088266372681,
      "epoch": 0.3884934756820878,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004990509620867997,
      "loss": 5.7112,
      "mean_token_accuracy": 0.16412873268127443,
      "num_tokens": 8986012.0,
      "step": 3930
    },
    {
      "entropy": 5.720556116104126,
      "epoch": 0.3889877421905892,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004990477214675868,
      "loss": 5.694,
      "mean_token_accuracy": 0.16026492416858673,
      "num_tokens": 8996759.0,
      "step": 3935
    },
    {
      "entropy": 5.849727296829224,
      "epoch": 0.38948200869909055,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004990444753367653,
      "loss": 5.829,
      "mean_token_accuracy": 0.15216880887746811,
      "num_tokens": 9007215.0,
      "step": 3940
    },
    {
      "entropy": 5.791629219055176,
      "epoch": 0.3899762752075919,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004990412236944149,
      "loss": 5.7292,
      "mean_token_accuracy": 0.17175416946411132,
      "num_tokens": 9018281.0,
      "step": 3945
    },
    {
      "entropy": 5.912288618087769,
      "epoch": 0.3904705417160933,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004990379665406157,
      "loss": 5.8764,
      "mean_token_accuracy": 0.15223623141646386,
      "num_tokens": 9029312.0,
      "step": 3950
    },
    {
      "entropy": 5.94501895904541,
      "epoch": 0.39096480822459473,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004990347038754477,
      "loss": 5.8566,
      "mean_token_accuracy": 0.14973756968975066,
      "num_tokens": 9040273.0,
      "step": 3955
    },
    {
      "entropy": 5.866108179092407,
      "epoch": 0.3914590747330961,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004990314356989912,
      "loss": 5.7922,
      "mean_token_accuracy": 0.15092711076140403,
      "num_tokens": 9052571.0,
      "step": 3960
    },
    {
      "entropy": 5.806940126419067,
      "epoch": 0.3919533412415975,
      "grad_norm": 1.5,
      "learning_rate": 0.0004990281620113268,
      "loss": 5.7422,
      "mean_token_accuracy": 0.15850640684366227,
      "num_tokens": 9063531.0,
      "step": 3965
    },
    {
      "entropy": 5.925866508483887,
      "epoch": 0.39244760775009885,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004990248828125347,
      "loss": 5.7836,
      "mean_token_accuracy": 0.15395828783512117,
      "num_tokens": 9075404.0,
      "step": 3970
    },
    {
      "entropy": 5.757915782928467,
      "epoch": 0.3929418742586002,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004990215981026958,
      "loss": 5.8031,
      "mean_token_accuracy": 0.15948758646845818,
      "num_tokens": 9086570.0,
      "step": 3975
    },
    {
      "entropy": 5.876324081420899,
      "epoch": 0.3934361407671016,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004990183078818908,
      "loss": 5.6828,
      "mean_token_accuracy": 0.1598816230893135,
      "num_tokens": 9097993.0,
      "step": 3980
    },
    {
      "entropy": 5.838431692123413,
      "epoch": 0.393930407275603,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004990150121502007,
      "loss": 5.7628,
      "mean_token_accuracy": 0.16408235877752303,
      "num_tokens": 9108783.0,
      "step": 3985
    },
    {
      "entropy": 5.788733673095703,
      "epoch": 0.3944246737841044,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004990117109077066,
      "loss": 5.8128,
      "mean_token_accuracy": 0.15555327832698823,
      "num_tokens": 9120039.0,
      "step": 3990
    },
    {
      "entropy": 5.8930316925048825,
      "epoch": 0.39491894029260577,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004990084041544895,
      "loss": 5.7652,
      "mean_token_accuracy": 0.15779706388711928,
      "num_tokens": 9131543.0,
      "step": 3995
    },
    {
      "entropy": 5.815483951568604,
      "epoch": 0.39541320680110714,
      "grad_norm": 1.078125,
      "learning_rate": 0.000499005091890631,
      "loss": 5.7806,
      "mean_token_accuracy": 0.15455961227416992,
      "num_tokens": 9142968.0,
      "step": 4000
    },
    {
      "entropy": 5.782220888137817,
      "epoch": 0.3959074733096085,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004990017741162125,
      "loss": 5.8476,
      "mean_token_accuracy": 0.15507417023181916,
      "num_tokens": 9155705.0,
      "step": 4005
    },
    {
      "entropy": 5.916092014312744,
      "epoch": 0.39640173981810994,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004989984508313156,
      "loss": 5.8555,
      "mean_token_accuracy": 0.14767262265086173,
      "num_tokens": 9167575.0,
      "step": 4010
    },
    {
      "entropy": 5.864837169647217,
      "epoch": 0.3968960063266113,
      "grad_norm": 1.203125,
      "learning_rate": 0.000498995122036022,
      "loss": 5.8303,
      "mean_token_accuracy": 0.15483421534299852,
      "num_tokens": 9178493.0,
      "step": 4015
    },
    {
      "entropy": 5.9377106666564945,
      "epoch": 0.3973902728351127,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004989917877304138,
      "loss": 5.8824,
      "mean_token_accuracy": 0.1505347326397896,
      "num_tokens": 9189677.0,
      "step": 4020
    },
    {
      "entropy": 5.827444219589234,
      "epoch": 0.39788453934361406,
      "grad_norm": 1.25,
      "learning_rate": 0.0004989884479145727,
      "loss": 5.7078,
      "mean_token_accuracy": 0.15970520824193954,
      "num_tokens": 9199179.0,
      "step": 4025
    },
    {
      "entropy": 5.905575847625732,
      "epoch": 0.39837880585211544,
      "grad_norm": 1.125,
      "learning_rate": 0.0004989851025885812,
      "loss": 5.778,
      "mean_token_accuracy": 0.15371362268924713,
      "num_tokens": 9210689.0,
      "step": 4030
    },
    {
      "entropy": 5.812503290176392,
      "epoch": 0.39887307236061686,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004989817517525212,
      "loss": 5.7441,
      "mean_token_accuracy": 0.15978539288043975,
      "num_tokens": 9221710.0,
      "step": 4035
    },
    {
      "entropy": 5.845702409744263,
      "epoch": 0.39936733886911824,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004989783954064755,
      "loss": 5.84,
      "mean_token_accuracy": 0.15523956567049027,
      "num_tokens": 9234079.0,
      "step": 4040
    },
    {
      "entropy": 5.842787218093872,
      "epoch": 0.3998616053776196,
      "grad_norm": 1.125,
      "learning_rate": 0.0004989750335505265,
      "loss": 5.7661,
      "mean_token_accuracy": 0.15682925432920455,
      "num_tokens": 9245050.0,
      "step": 4045
    },
    {
      "entropy": 5.9115033626556395,
      "epoch": 0.400355871886121,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004989716661847571,
      "loss": 5.8554,
      "mean_token_accuracy": 0.15244610458612443,
      "num_tokens": 9256455.0,
      "step": 4050
    },
    {
      "entropy": 5.779877328872681,
      "epoch": 0.40085013839462236,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004989682933092497,
      "loss": 5.7513,
      "mean_token_accuracy": 0.16125411689281463,
      "num_tokens": 9267722.0,
      "step": 4055
    },
    {
      "entropy": 5.817647457122803,
      "epoch": 0.4013444049031238,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004989649149240877,
      "loss": 5.7853,
      "mean_token_accuracy": 0.15072275847196578,
      "num_tokens": 9277533.0,
      "step": 4060
    },
    {
      "entropy": 5.887001609802246,
      "epoch": 0.40183867141162516,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000498961531029354,
      "loss": 5.8136,
      "mean_token_accuracy": 0.16053701788187028,
      "num_tokens": 9290052.0,
      "step": 4065
    },
    {
      "entropy": 5.850340270996094,
      "epoch": 0.40233293792012653,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000498958141625132,
      "loss": 5.8021,
      "mean_token_accuracy": 0.15744038969278334,
      "num_tokens": 9302201.0,
      "step": 4070
    },
    {
      "entropy": 5.914964580535889,
      "epoch": 0.4028272044286279,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004989547467115047,
      "loss": 5.8788,
      "mean_token_accuracy": 0.1513068199157715,
      "num_tokens": 9313215.0,
      "step": 4075
    },
    {
      "entropy": 5.910677099227906,
      "epoch": 0.4033214709371293,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004989513462885561,
      "loss": 5.917,
      "mean_token_accuracy": 0.145223967730999,
      "num_tokens": 9324098.0,
      "step": 4080
    },
    {
      "entropy": 5.964850568771363,
      "epoch": 0.4038157374456307,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004989479403563696,
      "loss": 5.7984,
      "mean_token_accuracy": 0.15427584648132325,
      "num_tokens": 9336048.0,
      "step": 4085
    },
    {
      "entropy": 5.874447202682495,
      "epoch": 0.4043100039541321,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004989445289150289,
      "loss": 5.7562,
      "mean_token_accuracy": 0.15428639203310013,
      "num_tokens": 9347723.0,
      "step": 4090
    },
    {
      "entropy": 5.81725811958313,
      "epoch": 0.40480427046263345,
      "grad_norm": 1.0,
      "learning_rate": 0.0004989411119646183,
      "loss": 5.802,
      "mean_token_accuracy": 0.15105294287204743,
      "num_tokens": 9360588.0,
      "step": 4095
    },
    {
      "entropy": 5.881053924560547,
      "epoch": 0.4052985369711348,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004989376895052214,
      "loss": 5.7714,
      "mean_token_accuracy": 0.1561678320169449,
      "num_tokens": 9371798.0,
      "step": 4100
    },
    {
      "entropy": 5.816941499710083,
      "epoch": 0.4057928034796362,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004989342615369226,
      "loss": 5.7775,
      "mean_token_accuracy": 0.16338589042425156,
      "num_tokens": 9382421.0,
      "step": 4105
    },
    {
      "entropy": 5.828975009918213,
      "epoch": 0.40628706998813763,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004989308280598063,
      "loss": 5.7102,
      "mean_token_accuracy": 0.16223904639482498,
      "num_tokens": 9392832.0,
      "step": 4110
    },
    {
      "entropy": 5.789451217651367,
      "epoch": 0.406781336496639,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004989273890739569,
      "loss": 5.7038,
      "mean_token_accuracy": 0.16221638768911362,
      "num_tokens": 9405309.0,
      "step": 4115
    },
    {
      "entropy": 5.795983219146729,
      "epoch": 0.4072756030051404,
      "grad_norm": 1.125,
      "learning_rate": 0.000498923944579459,
      "loss": 5.7135,
      "mean_token_accuracy": 0.16932716518640517,
      "num_tokens": 9416707.0,
      "step": 4120
    },
    {
      "entropy": 5.821607875823974,
      "epoch": 0.40776986951364175,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004989204945763972,
      "loss": 5.8453,
      "mean_token_accuracy": 0.1561842828989029,
      "num_tokens": 9427684.0,
      "step": 4125
    },
    {
      "entropy": 5.83395471572876,
      "epoch": 0.4082641360221431,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004989170390648567,
      "loss": 5.8027,
      "mean_token_accuracy": 0.15728633105754852,
      "num_tokens": 9439073.0,
      "step": 4130
    },
    {
      "entropy": 5.800190210342407,
      "epoch": 0.40875840253064455,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004989135780449222,
      "loss": 5.7427,
      "mean_token_accuracy": 0.15998244285583496,
      "num_tokens": 9450266.0,
      "step": 4135
    },
    {
      "entropy": 5.776017856597901,
      "epoch": 0.4092526690391459,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004989101115166789,
      "loss": 5.706,
      "mean_token_accuracy": 0.16565122753381728,
      "num_tokens": 9462662.0,
      "step": 4140
    },
    {
      "entropy": 5.86929087638855,
      "epoch": 0.4097469355476473,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004989066394802122,
      "loss": 5.8024,
      "mean_token_accuracy": 0.15239896178245543,
      "num_tokens": 9474338.0,
      "step": 4145
    },
    {
      "entropy": 5.845749139785767,
      "epoch": 0.41024120205614867,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004989031619356074,
      "loss": 5.7838,
      "mean_token_accuracy": 0.16020090132951736,
      "num_tokens": 9486353.0,
      "step": 4150
    },
    {
      "entropy": 5.8469836711883545,
      "epoch": 0.41073546856465004,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00049889967888295,
      "loss": 5.7535,
      "mean_token_accuracy": 0.1544842854142189,
      "num_tokens": 9497464.0,
      "step": 4155
    },
    {
      "entropy": 5.81313943862915,
      "epoch": 0.41122973507315147,
      "grad_norm": 1.125,
      "learning_rate": 0.0004988961903223258,
      "loss": 5.7288,
      "mean_token_accuracy": 0.1628125235438347,
      "num_tokens": 9508037.0,
      "step": 4160
    },
    {
      "entropy": 5.773452091217041,
      "epoch": 0.41172400158165284,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004988926962538207,
      "loss": 5.7676,
      "mean_token_accuracy": 0.15996645092964173,
      "num_tokens": 9518653.0,
      "step": 4165
    },
    {
      "entropy": 5.854199171066284,
      "epoch": 0.4122182680901542,
      "grad_norm": 1.25,
      "learning_rate": 0.0004988891966775204,
      "loss": 5.7609,
      "mean_token_accuracy": 0.16049931049346924,
      "num_tokens": 9529225.0,
      "step": 4170
    },
    {
      "entropy": 5.8694295406341555,
      "epoch": 0.4127125345986556,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004988856915935112,
      "loss": 5.7343,
      "mean_token_accuracy": 0.15607544779777527,
      "num_tokens": 9541010.0,
      "step": 4175
    },
    {
      "entropy": 5.747597646713257,
      "epoch": 0.41320680110715696,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004988821810018791,
      "loss": 5.8116,
      "mean_token_accuracy": 0.15240168645977975,
      "num_tokens": 9552061.0,
      "step": 4180
    },
    {
      "entropy": 5.953191566467285,
      "epoch": 0.41370106761565834,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004988786649027108,
      "loss": 5.9074,
      "mean_token_accuracy": 0.14590952843427657,
      "num_tokens": 9563970.0,
      "step": 4185
    },
    {
      "entropy": 5.872267389297486,
      "epoch": 0.41419533412415976,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004988751432960926,
      "loss": 5.6962,
      "mean_token_accuracy": 0.16297404766082763,
      "num_tokens": 9575381.0,
      "step": 4190
    },
    {
      "entropy": 5.843984842300415,
      "epoch": 0.41468960063266114,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004988716161821111,
      "loss": 5.7267,
      "mean_token_accuracy": 0.16292373985052108,
      "num_tokens": 9586821.0,
      "step": 4195
    },
    {
      "entropy": 5.843246555328369,
      "epoch": 0.4151838671411625,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004988680835608531,
      "loss": 5.8082,
      "mean_token_accuracy": 0.15058472603559495,
      "num_tokens": 9599580.0,
      "step": 4200
    },
    {
      "entropy": 5.7772629737854,
      "epoch": 0.4156781336496639,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004988645454324055,
      "loss": 5.8068,
      "mean_token_accuracy": 0.15320300832390785,
      "num_tokens": 9611495.0,
      "step": 4205
    },
    {
      "entropy": 5.92116961479187,
      "epoch": 0.41617240015816526,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004988610017968554,
      "loss": 5.7728,
      "mean_token_accuracy": 0.15595485121011735,
      "num_tokens": 9623713.0,
      "step": 4210
    },
    {
      "entropy": 5.768342113494873,
      "epoch": 0.4166666666666667,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004988574526542899,
      "loss": 5.7569,
      "mean_token_accuracy": 0.1648036479949951,
      "num_tokens": 9633922.0,
      "step": 4215
    },
    {
      "entropy": 5.831979560852051,
      "epoch": 0.41716093317516806,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004988538980047963,
      "loss": 5.8424,
      "mean_token_accuracy": 0.15677017271518706,
      "num_tokens": 9645641.0,
      "step": 4220
    },
    {
      "entropy": 5.851514863967895,
      "epoch": 0.41765519968366943,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004988503378484622,
      "loss": 5.7467,
      "mean_token_accuracy": 0.16601389199495314,
      "num_tokens": 9656927.0,
      "step": 4225
    },
    {
      "entropy": 5.79203519821167,
      "epoch": 0.4181494661921708,
      "grad_norm": 1.25,
      "learning_rate": 0.000498846772185375,
      "loss": 5.7172,
      "mean_token_accuracy": 0.1635448768734932,
      "num_tokens": 9669148.0,
      "step": 4230
    },
    {
      "entropy": 5.891840267181396,
      "epoch": 0.4186437327006722,
      "grad_norm": 1.125,
      "learning_rate": 0.0004988432010156224,
      "loss": 5.838,
      "mean_token_accuracy": 0.1492895156145096,
      "num_tokens": 9680620.0,
      "step": 4235
    },
    {
      "entropy": 5.9005063533782955,
      "epoch": 0.4191379992091736,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004988396243392924,
      "loss": 5.7901,
      "mean_token_accuracy": 0.15724762678146362,
      "num_tokens": 9690984.0,
      "step": 4240
    },
    {
      "entropy": 5.832825088500977,
      "epoch": 0.419632265717675,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000498836042156473,
      "loss": 5.768,
      "mean_token_accuracy": 0.16585759073495865,
      "num_tokens": 9701561.0,
      "step": 4245
    },
    {
      "entropy": 5.790317058563232,
      "epoch": 0.42012653222617635,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004988324544672521,
      "loss": 5.7601,
      "mean_token_accuracy": 0.1544044718146324,
      "num_tokens": 9712780.0,
      "step": 4250
    },
    {
      "entropy": 5.848903608322144,
      "epoch": 0.4206207987346777,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004988288612717182,
      "loss": 5.7547,
      "mean_token_accuracy": 0.1562611922621727,
      "num_tokens": 9724547.0,
      "step": 4255
    },
    {
      "entropy": 5.8364794731140135,
      "epoch": 0.4211150652431791,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004988252625699594,
      "loss": 5.8111,
      "mean_token_accuracy": 0.15629005581140518,
      "num_tokens": 9736897.0,
      "step": 4260
    },
    {
      "entropy": 5.8446488857269285,
      "epoch": 0.4216093317516805,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004988216583620647,
      "loss": 5.7584,
      "mean_token_accuracy": 0.16204439848661423,
      "num_tokens": 9748214.0,
      "step": 4265
    },
    {
      "entropy": 5.886655187606811,
      "epoch": 0.4221035982601819,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004988180486481223,
      "loss": 5.7894,
      "mean_token_accuracy": 0.15823814868927003,
      "num_tokens": 9759602.0,
      "step": 4270
    },
    {
      "entropy": 5.827601718902588,
      "epoch": 0.4225978647686833,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004988144334282211,
      "loss": 5.7425,
      "mean_token_accuracy": 0.1589537337422371,
      "num_tokens": 9771645.0,
      "step": 4275
    },
    {
      "entropy": 5.853240966796875,
      "epoch": 0.42309213127718465,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004988108127024503,
      "loss": 5.7505,
      "mean_token_accuracy": 0.1587558016180992,
      "num_tokens": 9783504.0,
      "step": 4280
    },
    {
      "entropy": 5.891699171066284,
      "epoch": 0.423586397785686,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004988071864708987,
      "loss": 5.8804,
      "mean_token_accuracy": 0.15425789207220078,
      "num_tokens": 9795572.0,
      "step": 4285
    },
    {
      "entropy": 5.83182110786438,
      "epoch": 0.42408066429418745,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004988035547336555,
      "loss": 5.7336,
      "mean_token_accuracy": 0.15293268710374833,
      "num_tokens": 9806398.0,
      "step": 4290
    },
    {
      "entropy": 5.82321310043335,
      "epoch": 0.4245749308026888,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004987999174908104,
      "loss": 5.7342,
      "mean_token_accuracy": 0.1546010985970497,
      "num_tokens": 9817171.0,
      "step": 4295
    },
    {
      "entropy": 5.818284416198731,
      "epoch": 0.4250691973111902,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004987962747424524,
      "loss": 5.7763,
      "mean_token_accuracy": 0.15528347045183183,
      "num_tokens": 9827535.0,
      "step": 4300
    },
    {
      "entropy": 5.79456467628479,
      "epoch": 0.42556346381969157,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004987926264886713,
      "loss": 5.7551,
      "mean_token_accuracy": 0.1570479989051819,
      "num_tokens": 9839619.0,
      "step": 4305
    },
    {
      "entropy": 5.840618705749511,
      "epoch": 0.42605773032819294,
      "grad_norm": 1.28125,
      "learning_rate": 0.000498788972729557,
      "loss": 5.7544,
      "mean_token_accuracy": 0.15230125635862352,
      "num_tokens": 9850100.0,
      "step": 4310
    },
    {
      "entropy": 5.917747354507446,
      "epoch": 0.42655199683669437,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004987853134651992,
      "loss": 5.8327,
      "mean_token_accuracy": 0.14419816955924034,
      "num_tokens": 9862228.0,
      "step": 4315
    },
    {
      "entropy": 5.803866004943847,
      "epoch": 0.42704626334519574,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000498781648695688,
      "loss": 5.7229,
      "mean_token_accuracy": 0.1623837560415268,
      "num_tokens": 9873080.0,
      "step": 4320
    },
    {
      "entropy": 5.791725206375122,
      "epoch": 0.4275405298536971,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004987779784211135,
      "loss": 5.7442,
      "mean_token_accuracy": 0.15227282643318177,
      "num_tokens": 9884735.0,
      "step": 4325
    },
    {
      "entropy": 5.879214715957642,
      "epoch": 0.4280347963621985,
      "grad_norm": 1.234375,
      "learning_rate": 0.000498774302641566,
      "loss": 5.7641,
      "mean_token_accuracy": 0.15441785752773285,
      "num_tokens": 9895018.0,
      "step": 4330
    },
    {
      "entropy": 5.850545740127563,
      "epoch": 0.42852906287069986,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004987706213571359,
      "loss": 5.7371,
      "mean_token_accuracy": 0.1566600888967514,
      "num_tokens": 9906838.0,
      "step": 4335
    },
    {
      "entropy": 5.844648790359497,
      "epoch": 0.4290233293792013,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004987669345679139,
      "loss": 5.7701,
      "mean_token_accuracy": 0.1565292850136757,
      "num_tokens": 9919556.0,
      "step": 4340
    },
    {
      "entropy": 5.781128025054931,
      "epoch": 0.42951759588770266,
      "grad_norm": 1.25,
      "learning_rate": 0.0004987632422739904,
      "loss": 5.6781,
      "mean_token_accuracy": 0.15627526938915254,
      "num_tokens": 9929783.0,
      "step": 4345
    },
    {
      "entropy": 5.772721195220948,
      "epoch": 0.43001186239620404,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004987595444754567,
      "loss": 5.7391,
      "mean_token_accuracy": 0.15912755578756332,
      "num_tokens": 9940894.0,
      "step": 4350
    },
    {
      "entropy": 5.80626540184021,
      "epoch": 0.4305061289047054,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004987558411724032,
      "loss": 5.718,
      "mean_token_accuracy": 0.16627187728881837,
      "num_tokens": 9951973.0,
      "step": 4355
    },
    {
      "entropy": 5.9029295444488525,
      "epoch": 0.4310003954132068,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004987521323649215,
      "loss": 5.7694,
      "mean_token_accuracy": 0.15462518632411956,
      "num_tokens": 9963304.0,
      "step": 4360
    },
    {
      "entropy": 5.838963317871094,
      "epoch": 0.4314946619217082,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004987484180531026,
      "loss": 5.7812,
      "mean_token_accuracy": 0.16214534044265747,
      "num_tokens": 9973981.0,
      "step": 4365
    },
    {
      "entropy": 5.8302850246429445,
      "epoch": 0.4319889284302096,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004987446982370379,
      "loss": 5.8036,
      "mean_token_accuracy": 0.1544625535607338,
      "num_tokens": 9986461.0,
      "step": 4370
    },
    {
      "entropy": 5.8808770179748535,
      "epoch": 0.43248319493871096,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000498740972916819,
      "loss": 5.7715,
      "mean_token_accuracy": 0.15348488688468934,
      "num_tokens": 9997973.0,
      "step": 4375
    },
    {
      "entropy": 5.828121280670166,
      "epoch": 0.43297746144721233,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004987372420925373,
      "loss": 5.78,
      "mean_token_accuracy": 0.15659529715776443,
      "num_tokens": 10009572.0,
      "step": 4380
    },
    {
      "entropy": 5.816518115997314,
      "epoch": 0.4334717279557137,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004987335057642847,
      "loss": 5.6502,
      "mean_token_accuracy": 0.1678347870707512,
      "num_tokens": 10020437.0,
      "step": 4385
    },
    {
      "entropy": 5.809616422653198,
      "epoch": 0.43396599446421513,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004987297639321533,
      "loss": 5.7808,
      "mean_token_accuracy": 0.15647988617420197,
      "num_tokens": 10031827.0,
      "step": 4390
    },
    {
      "entropy": 5.821065092086792,
      "epoch": 0.4344602609727165,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004987260165962349,
      "loss": 5.7813,
      "mean_token_accuracy": 0.15543572455644608,
      "num_tokens": 10043833.0,
      "step": 4395
    },
    {
      "entropy": 5.8089728355407715,
      "epoch": 0.4349545274812179,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004987222637566218,
      "loss": 5.7225,
      "mean_token_accuracy": 0.1572750225663185,
      "num_tokens": 10054645.0,
      "step": 4400
    },
    {
      "entropy": 5.7818284034729,
      "epoch": 0.43544879398971925,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004987185054134064,
      "loss": 5.6724,
      "mean_token_accuracy": 0.1580430954694748,
      "num_tokens": 10065178.0,
      "step": 4405
    },
    {
      "entropy": 5.825777530670166,
      "epoch": 0.4359430604982206,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000498714741566681,
      "loss": 5.706,
      "mean_token_accuracy": 0.16232621520757676,
      "num_tokens": 10076701.0,
      "step": 4410
    },
    {
      "entropy": 5.735870885848999,
      "epoch": 0.436437327006722,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004987109722165383,
      "loss": 5.7263,
      "mean_token_accuracy": 0.15729656368494033,
      "num_tokens": 10087959.0,
      "step": 4415
    },
    {
      "entropy": 5.88615665435791,
      "epoch": 0.4369315935152234,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004987071973630708,
      "loss": 5.7156,
      "mean_token_accuracy": 0.15740345194935798,
      "num_tokens": 10099463.0,
      "step": 4420
    },
    {
      "entropy": 5.822025537490845,
      "epoch": 0.4374258600237248,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004987034170063717,
      "loss": 5.8005,
      "mean_token_accuracy": 0.15701815336942673,
      "num_tokens": 10111384.0,
      "step": 4425
    },
    {
      "entropy": 5.821445465087891,
      "epoch": 0.4379201265322262,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004986996311465338,
      "loss": 5.7321,
      "mean_token_accuracy": 0.15710116550326347,
      "num_tokens": 10121838.0,
      "step": 4430
    },
    {
      "entropy": 5.8637402057647705,
      "epoch": 0.43841439304072755,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004986958397836503,
      "loss": 5.6928,
      "mean_token_accuracy": 0.1640600949525833,
      "num_tokens": 10132225.0,
      "step": 4435
    },
    {
      "entropy": 5.729298543930054,
      "epoch": 0.4389086595492289,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004986920429178144,
      "loss": 5.6631,
      "mean_token_accuracy": 0.15749302357435227,
      "num_tokens": 10143114.0,
      "step": 4440
    },
    {
      "entropy": 5.893101644515991,
      "epoch": 0.43940292605773035,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004986882405491195,
      "loss": 5.8626,
      "mean_token_accuracy": 0.15157403349876403,
      "num_tokens": 10155057.0,
      "step": 4445
    },
    {
      "entropy": 5.75573582649231,
      "epoch": 0.4398971925662317,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004986844326776593,
      "loss": 5.6229,
      "mean_token_accuracy": 0.1684441089630127,
      "num_tokens": 10166816.0,
      "step": 4450
    },
    {
      "entropy": 5.734735536575317,
      "epoch": 0.4403914590747331,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004986806193035272,
      "loss": 5.695,
      "mean_token_accuracy": 0.1676469177007675,
      "num_tokens": 10179433.0,
      "step": 4455
    },
    {
      "entropy": 5.8267802715301515,
      "epoch": 0.44088572558323447,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004986768004268173,
      "loss": 5.693,
      "mean_token_accuracy": 0.16008566021919252,
      "num_tokens": 10190784.0,
      "step": 4460
    },
    {
      "entropy": 5.836114597320557,
      "epoch": 0.44137999209173584,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004986729760476234,
      "loss": 5.7523,
      "mean_token_accuracy": 0.15255020558834076,
      "num_tokens": 10201941.0,
      "step": 4465
    },
    {
      "entropy": 5.808204221725464,
      "epoch": 0.44187425860023727,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004986691461660394,
      "loss": 5.7176,
      "mean_token_accuracy": 0.16116027235984803,
      "num_tokens": 10214029.0,
      "step": 4470
    },
    {
      "entropy": 5.776810598373413,
      "epoch": 0.44236852510873864,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004986653107821599,
      "loss": 5.7602,
      "mean_token_accuracy": 0.16111526936292647,
      "num_tokens": 10225585.0,
      "step": 4475
    },
    {
      "entropy": 5.78863844871521,
      "epoch": 0.44286279161724,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004986614698960789,
      "loss": 5.7031,
      "mean_token_accuracy": 0.16227239221334458,
      "num_tokens": 10237753.0,
      "step": 4480
    },
    {
      "entropy": 5.912133312225341,
      "epoch": 0.4433570581257414,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004986576235078913,
      "loss": 5.9483,
      "mean_token_accuracy": 0.1438101887702942,
      "num_tokens": 10251522.0,
      "step": 4485
    },
    {
      "entropy": 5.831998538970947,
      "epoch": 0.44385132463424276,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004986537716176912,
      "loss": 5.6578,
      "mean_token_accuracy": 0.16271060109138488,
      "num_tokens": 10263319.0,
      "step": 4490
    },
    {
      "entropy": 5.715137338638305,
      "epoch": 0.4443455911427442,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004986499142255738,
      "loss": 5.7045,
      "mean_token_accuracy": 0.15843060836195946,
      "num_tokens": 10274899.0,
      "step": 4495
    },
    {
      "entropy": 5.836452913284302,
      "epoch": 0.44483985765124556,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004986460513316338,
      "loss": 5.676,
      "mean_token_accuracy": 0.16559335291385652,
      "num_tokens": 10286110.0,
      "step": 4500
    },
    {
      "entropy": 5.78662633895874,
      "epoch": 0.44533412415974694,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004986421829359662,
      "loss": 5.8238,
      "mean_token_accuracy": 0.15288708209991456,
      "num_tokens": 10298870.0,
      "step": 4505
    },
    {
      "entropy": 5.843674945831299,
      "epoch": 0.4458283906682483,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004986383090386661,
      "loss": 5.744,
      "mean_token_accuracy": 0.15693433284759523,
      "num_tokens": 10310326.0,
      "step": 4510
    },
    {
      "entropy": 5.83041410446167,
      "epoch": 0.4463226571767497,
      "grad_norm": 1.265625,
      "learning_rate": 0.000498634429639829,
      "loss": 5.6721,
      "mean_token_accuracy": 0.15878720134496688,
      "num_tokens": 10320899.0,
      "step": 4515
    },
    {
      "entropy": 5.792200517654419,
      "epoch": 0.4468169236852511,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004986305447395502,
      "loss": 5.7275,
      "mean_token_accuracy": 0.1623765215277672,
      "num_tokens": 10331669.0,
      "step": 4520
    },
    {
      "entropy": 5.829644346237183,
      "epoch": 0.4473111901937525,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004986266543379254,
      "loss": 5.7114,
      "mean_token_accuracy": 0.1589057058095932,
      "num_tokens": 10343120.0,
      "step": 4525
    },
    {
      "entropy": 5.772820949554443,
      "epoch": 0.44780545670225386,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004986227584350501,
      "loss": 5.7149,
      "mean_token_accuracy": 0.1589616447687149,
      "num_tokens": 10354806.0,
      "step": 4530
    },
    {
      "entropy": 5.848662900924682,
      "epoch": 0.44829972321075523,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004986188570310203,
      "loss": 5.7359,
      "mean_token_accuracy": 0.1575729265809059,
      "num_tokens": 10366453.0,
      "step": 4535
    },
    {
      "entropy": 5.86790361404419,
      "epoch": 0.4487939897192566,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004986149501259319,
      "loss": 5.8281,
      "mean_token_accuracy": 0.15670044720172882,
      "num_tokens": 10378095.0,
      "step": 4540
    },
    {
      "entropy": 5.732289028167725,
      "epoch": 0.44928825622775803,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004986110377198812,
      "loss": 5.6931,
      "mean_token_accuracy": 0.16421767175197602,
      "num_tokens": 10389620.0,
      "step": 4545
    },
    {
      "entropy": 5.868263101577758,
      "epoch": 0.4497825227362594,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000498607119812964,
      "loss": 5.8237,
      "mean_token_accuracy": 0.15202950835227966,
      "num_tokens": 10400166.0,
      "step": 4550
    },
    {
      "entropy": 5.809820365905762,
      "epoch": 0.4502767892447608,
      "grad_norm": 1.25,
      "learning_rate": 0.0004986031964052772,
      "loss": 5.7247,
      "mean_token_accuracy": 0.15717918127775193,
      "num_tokens": 10411116.0,
      "step": 4555
    },
    {
      "entropy": 5.741772174835205,
      "epoch": 0.45077105575326215,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004985992674969169,
      "loss": 5.7477,
      "mean_token_accuracy": 0.15264249593019485,
      "num_tokens": 10423484.0,
      "step": 4560
    },
    {
      "entropy": 5.902524757385254,
      "epoch": 0.4512653222617635,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004985953330879801,
      "loss": 5.813,
      "mean_token_accuracy": 0.14781858772039413,
      "num_tokens": 10435636.0,
      "step": 4565
    },
    {
      "entropy": 5.847655820846557,
      "epoch": 0.45175958877026495,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004985913931785633,
      "loss": 5.7489,
      "mean_token_accuracy": 0.1599886104464531,
      "num_tokens": 10447586.0,
      "step": 4570
    },
    {
      "entropy": 5.782388496398926,
      "epoch": 0.4522538552787663,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004985874477687636,
      "loss": 5.8058,
      "mean_token_accuracy": 0.1555984526872635,
      "num_tokens": 10460678.0,
      "step": 4575
    },
    {
      "entropy": 5.818998098373413,
      "epoch": 0.4527481217872677,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004985834968586779,
      "loss": 5.7685,
      "mean_token_accuracy": 0.1549448549747467,
      "num_tokens": 10472738.0,
      "step": 4580
    },
    {
      "entropy": 5.85989031791687,
      "epoch": 0.45324238829576907,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004985795404484037,
      "loss": 5.6418,
      "mean_token_accuracy": 0.16428616791963577,
      "num_tokens": 10483716.0,
      "step": 4585
    },
    {
      "entropy": 5.747600412368774,
      "epoch": 0.45373665480427045,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004985755785380379,
      "loss": 5.7724,
      "mean_token_accuracy": 0.15435491055250167,
      "num_tokens": 10495287.0,
      "step": 4590
    },
    {
      "entropy": 5.795434284210205,
      "epoch": 0.4542309213127719,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004985716111276781,
      "loss": 5.7235,
      "mean_token_accuracy": 0.16080528646707534,
      "num_tokens": 10506332.0,
      "step": 4595
    },
    {
      "entropy": 5.858145904541016,
      "epoch": 0.45472518782127325,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004985676382174223,
      "loss": 5.7538,
      "mean_token_accuracy": 0.15550379902124406,
      "num_tokens": 10517574.0,
      "step": 4600
    },
    {
      "entropy": 5.77792444229126,
      "epoch": 0.4552194543297746,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004985636598073677,
      "loss": 5.6777,
      "mean_token_accuracy": 0.1682252198457718,
      "num_tokens": 10528998.0,
      "step": 4605
    },
    {
      "entropy": 5.829964351654053,
      "epoch": 0.455713720838276,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004985596758976125,
      "loss": 5.7456,
      "mean_token_accuracy": 0.15128750652074813,
      "num_tokens": 10539669.0,
      "step": 4610
    },
    {
      "entropy": 5.807875728607177,
      "epoch": 0.45620798734677737,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004985556864882545,
      "loss": 5.7253,
      "mean_token_accuracy": 0.16209764927625656,
      "num_tokens": 10550005.0,
      "step": 4615
    },
    {
      "entropy": 5.745830821990967,
      "epoch": 0.45670225385527874,
      "grad_norm": 1.171875,
      "learning_rate": 0.000498551691579392,
      "loss": 5.6994,
      "mean_token_accuracy": 0.16218698173761367,
      "num_tokens": 10561122.0,
      "step": 4620
    },
    {
      "entropy": 5.817157936096192,
      "epoch": 0.45719652036378017,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004985476911711232,
      "loss": 5.6775,
      "mean_token_accuracy": 0.15966929942369462,
      "num_tokens": 10572081.0,
      "step": 4625
    },
    {
      "entropy": 5.656571245193481,
      "epoch": 0.45769078687228154,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004985436852635465,
      "loss": 5.6708,
      "mean_token_accuracy": 0.1661362901329994,
      "num_tokens": 10582297.0,
      "step": 4630
    },
    {
      "entropy": 5.8948314666748045,
      "epoch": 0.4581850533807829,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004985396738567604,
      "loss": 5.7712,
      "mean_token_accuracy": 0.15315597653388976,
      "num_tokens": 10593161.0,
      "step": 4635
    },
    {
      "entropy": 5.829736757278442,
      "epoch": 0.4586793198892843,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004985356569508638,
      "loss": 5.7647,
      "mean_token_accuracy": 0.1625633120536804,
      "num_tokens": 10603900.0,
      "step": 4640
    },
    {
      "entropy": 5.788417482376099,
      "epoch": 0.45917358639778566,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004985316345459551,
      "loss": 5.7583,
      "mean_token_accuracy": 0.16019802242517472,
      "num_tokens": 10614946.0,
      "step": 4645
    },
    {
      "entropy": 5.819199132919311,
      "epoch": 0.4596678529062871,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004985276066421338,
      "loss": 5.7609,
      "mean_token_accuracy": 0.1518936887383461,
      "num_tokens": 10625073.0,
      "step": 4650
    },
    {
      "entropy": 5.856818628311157,
      "epoch": 0.46016211941478846,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004985235732394986,
      "loss": 5.8623,
      "mean_token_accuracy": 0.15111786425113677,
      "num_tokens": 10636942.0,
      "step": 4655
    },
    {
      "entropy": 5.893058109283447,
      "epoch": 0.46065638592328984,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004985195343381487,
      "loss": 5.7983,
      "mean_token_accuracy": 0.1530800074338913,
      "num_tokens": 10647608.0,
      "step": 4660
    },
    {
      "entropy": 5.854513692855835,
      "epoch": 0.4611506524317912,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004985154899381837,
      "loss": 5.8287,
      "mean_token_accuracy": 0.15122355073690413,
      "num_tokens": 10659532.0,
      "step": 4665
    },
    {
      "entropy": 5.853507041931152,
      "epoch": 0.4616449189402926,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004985114400397029,
      "loss": 5.7571,
      "mean_token_accuracy": 0.16026072353124618,
      "num_tokens": 10670342.0,
      "step": 4670
    },
    {
      "entropy": 5.8415283203125,
      "epoch": 0.462139185448794,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004985073846428061,
      "loss": 5.7877,
      "mean_token_accuracy": 0.15555054545402527,
      "num_tokens": 10682984.0,
      "step": 4675
    },
    {
      "entropy": 5.841789770126343,
      "epoch": 0.4626334519572954,
      "grad_norm": 1.125,
      "learning_rate": 0.0004985033237475929,
      "loss": 5.773,
      "mean_token_accuracy": 0.15423207432031633,
      "num_tokens": 10694235.0,
      "step": 4680
    },
    {
      "entropy": 5.926882314682007,
      "epoch": 0.46312771846579676,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004984992573541633,
      "loss": 5.7837,
      "mean_token_accuracy": 0.15645679384469985,
      "num_tokens": 10704508.0,
      "step": 4685
    },
    {
      "entropy": 5.676226234436035,
      "epoch": 0.46362198497429813,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004984951854626173,
      "loss": 5.6549,
      "mean_token_accuracy": 0.1645615130662918,
      "num_tokens": 10716927.0,
      "step": 4690
    },
    {
      "entropy": 5.752277278900147,
      "epoch": 0.4641162514827995,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004984911080730551,
      "loss": 5.6912,
      "mean_token_accuracy": 0.16231297999620437,
      "num_tokens": 10728396.0,
      "step": 4695
    },
    {
      "entropy": 5.783680486679077,
      "epoch": 0.46461051799130093,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004984870251855769,
      "loss": 5.7673,
      "mean_token_accuracy": 0.16363271474838256,
      "num_tokens": 10739001.0,
      "step": 4700
    },
    {
      "entropy": 5.851905822753906,
      "epoch": 0.4651047844998023,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004984829368002831,
      "loss": 5.7344,
      "mean_token_accuracy": 0.15897003412246705,
      "num_tokens": 10748768.0,
      "step": 4705
    },
    {
      "entropy": 5.860234689712525,
      "epoch": 0.4655990510083037,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004984788429172745,
      "loss": 5.7698,
      "mean_token_accuracy": 0.1561150312423706,
      "num_tokens": 10760582.0,
      "step": 4710
    },
    {
      "entropy": 5.760329294204712,
      "epoch": 0.46609331751680505,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004984747435366517,
      "loss": 5.6788,
      "mean_token_accuracy": 0.16664675623178482,
      "num_tokens": 10772537.0,
      "step": 4715
    },
    {
      "entropy": 5.772965812683106,
      "epoch": 0.4665875840253064,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004984706386585156,
      "loss": 5.6206,
      "mean_token_accuracy": 0.168980173766613,
      "num_tokens": 10783740.0,
      "step": 4720
    },
    {
      "entropy": 5.840044403076172,
      "epoch": 0.46708185053380785,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000498466528282967,
      "loss": 5.77,
      "mean_token_accuracy": 0.15619183331727982,
      "num_tokens": 10794793.0,
      "step": 4725
    },
    {
      "entropy": 5.840643644332886,
      "epoch": 0.4675761170423092,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004984624124101072,
      "loss": 5.7317,
      "mean_token_accuracy": 0.1579558730125427,
      "num_tokens": 10805944.0,
      "step": 4730
    },
    {
      "entropy": 5.8369344711303714,
      "epoch": 0.4680703835508106,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004984582910400374,
      "loss": 5.7556,
      "mean_token_accuracy": 0.15298434495925903,
      "num_tokens": 10818362.0,
      "step": 4735
    },
    {
      "entropy": 5.780828523635864,
      "epoch": 0.46856465005931197,
      "grad_norm": 1.109375,
      "learning_rate": 0.000498454164172859,
      "loss": 5.7502,
      "mean_token_accuracy": 0.16281691044569016,
      "num_tokens": 10831306.0,
      "step": 4740
    },
    {
      "entropy": 5.807979011535645,
      "epoch": 0.46905891656781334,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004984500318086733,
      "loss": 5.7066,
      "mean_token_accuracy": 0.16005800068378448,
      "num_tokens": 10841671.0,
      "step": 4745
    },
    {
      "entropy": 5.77095742225647,
      "epoch": 0.4695531830763148,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004984458939475824,
      "loss": 5.6916,
      "mean_token_accuracy": 0.1572381556034088,
      "num_tokens": 10852754.0,
      "step": 4750
    },
    {
      "entropy": 5.815678262710572,
      "epoch": 0.47004744958481615,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004984417505896877,
      "loss": 5.7662,
      "mean_token_accuracy": 0.156097312271595,
      "num_tokens": 10864812.0,
      "step": 4755
    },
    {
      "entropy": 5.857409429550171,
      "epoch": 0.4705417160933175,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004984376017350913,
      "loss": 5.7473,
      "mean_token_accuracy": 0.1609667122364044,
      "num_tokens": 10875967.0,
      "step": 4760
    },
    {
      "entropy": 5.79653754234314,
      "epoch": 0.4710359826018189,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004984334473838953,
      "loss": 5.7227,
      "mean_token_accuracy": 0.16150201112031937,
      "num_tokens": 10889575.0,
      "step": 4765
    },
    {
      "entropy": 5.82901701927185,
      "epoch": 0.47153024911032027,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004984292875362018,
      "loss": 5.804,
      "mean_token_accuracy": 0.1580755218863487,
      "num_tokens": 10900751.0,
      "step": 4770
    },
    {
      "entropy": 5.878472757339478,
      "epoch": 0.4720245156188217,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004984251221921131,
      "loss": 5.7473,
      "mean_token_accuracy": 0.15800192058086396,
      "num_tokens": 10912618.0,
      "step": 4775
    },
    {
      "entropy": 5.765990495681763,
      "epoch": 0.47251878212732307,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004984209513517317,
      "loss": 5.7156,
      "mean_token_accuracy": 0.15646711066365243,
      "num_tokens": 10925286.0,
      "step": 4780
    },
    {
      "entropy": 5.829292821884155,
      "epoch": 0.47301304863582444,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004984167750151603,
      "loss": 5.7242,
      "mean_token_accuracy": 0.16060231178998946,
      "num_tokens": 10936033.0,
      "step": 4785
    },
    {
      "entropy": 5.779935884475708,
      "epoch": 0.4735073151443258,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004984125931825018,
      "loss": 5.6637,
      "mean_token_accuracy": 0.16456516981124877,
      "num_tokens": 10947055.0,
      "step": 4790
    },
    {
      "entropy": 5.793975830078125,
      "epoch": 0.4740015816528272,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004984084058538585,
      "loss": 5.7137,
      "mean_token_accuracy": 0.1572008863091469,
      "num_tokens": 10957468.0,
      "step": 4795
    },
    {
      "entropy": 5.861984443664551,
      "epoch": 0.4744958481613286,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004984042130293339,
      "loss": 5.8362,
      "mean_token_accuracy": 0.1582918718457222,
      "num_tokens": 10970273.0,
      "step": 4800
    },
    {
      "entropy": 5.86493616104126,
      "epoch": 0.47499011466983,
      "grad_norm": 1.375,
      "learning_rate": 0.0004984000147090311,
      "loss": 5.8211,
      "mean_token_accuracy": 0.1545470640063286,
      "num_tokens": 10981842.0,
      "step": 4805
    },
    {
      "entropy": 5.888898468017578,
      "epoch": 0.47548438117833136,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004983958108930531,
      "loss": 5.7169,
      "mean_token_accuracy": 0.15943304374814032,
      "num_tokens": 10993615.0,
      "step": 4810
    },
    {
      "entropy": 5.83760347366333,
      "epoch": 0.47597864768683273,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004983916015815036,
      "loss": 5.8064,
      "mean_token_accuracy": 0.15529759526252745,
      "num_tokens": 11004451.0,
      "step": 4815
    },
    {
      "entropy": 5.901088237762451,
      "epoch": 0.4764729141953341,
      "grad_norm": 1.53125,
      "learning_rate": 0.000498387386774486,
      "loss": 5.745,
      "mean_token_accuracy": 0.15757741779088974,
      "num_tokens": 11014834.0,
      "step": 4820
    },
    {
      "entropy": 5.731997919082642,
      "epoch": 0.4769671807038355,
      "grad_norm": 1.375,
      "learning_rate": 0.000498383166472104,
      "loss": 5.6858,
      "mean_token_accuracy": 0.15810937881469728,
      "num_tokens": 11026116.0,
      "step": 4825
    },
    {
      "entropy": 5.807245063781738,
      "epoch": 0.4774614472123369,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004983789406744615,
      "loss": 5.7472,
      "mean_token_accuracy": 0.15772093832492828,
      "num_tokens": 11038780.0,
      "step": 4830
    },
    {
      "entropy": 5.878645515441894,
      "epoch": 0.4779557137208383,
      "grad_norm": 1.375,
      "learning_rate": 0.0004983747093816625,
      "loss": 5.7909,
      "mean_token_accuracy": 0.1571904167532921,
      "num_tokens": 11049824.0,
      "step": 4835
    },
    {
      "entropy": 5.7510171890258786,
      "epoch": 0.47844998022933966,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004983704725938108,
      "loss": 5.6787,
      "mean_token_accuracy": 0.1669359490275383,
      "num_tokens": 11060863.0,
      "step": 4840
    },
    {
      "entropy": 5.7988502979278564,
      "epoch": 0.47894424673784103,
      "grad_norm": 1.28125,
      "learning_rate": 0.000498366230311011,
      "loss": 5.6829,
      "mean_token_accuracy": 0.1632984146475792,
      "num_tokens": 11072533.0,
      "step": 4845
    },
    {
      "entropy": 5.72292914390564,
      "epoch": 0.4794385132463424,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004983619825333671,
      "loss": 5.6217,
      "mean_token_accuracy": 0.1602368801832199,
      "num_tokens": 11083627.0,
      "step": 4850
    },
    {
      "entropy": 5.8456309795379635,
      "epoch": 0.47993277975484383,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004983577292609839,
      "loss": 5.8346,
      "mean_token_accuracy": 0.14952256679534912,
      "num_tokens": 11096421.0,
      "step": 4855
    },
    {
      "entropy": 5.796829748153686,
      "epoch": 0.4804270462633452,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004983534704939659,
      "loss": 5.6678,
      "mean_token_accuracy": 0.15885301381349565,
      "num_tokens": 11109053.0,
      "step": 4860
    },
    {
      "entropy": 5.730153894424438,
      "epoch": 0.4809213127718466,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004983492062324179,
      "loss": 5.6214,
      "mean_token_accuracy": 0.16144478619098662,
      "num_tokens": 11120431.0,
      "step": 4865
    },
    {
      "entropy": 5.791682338714599,
      "epoch": 0.48141557928034795,
      "grad_norm": 1.375,
      "learning_rate": 0.0004983449364764447,
      "loss": 5.7241,
      "mean_token_accuracy": 0.1602831542491913,
      "num_tokens": 11132767.0,
      "step": 4870
    },
    {
      "entropy": 5.746629905700684,
      "epoch": 0.4819098457888493,
      "grad_norm": 1.375,
      "learning_rate": 0.0004983406612261514,
      "loss": 5.7184,
      "mean_token_accuracy": 0.15982515960931779,
      "num_tokens": 11144508.0,
      "step": 4875
    },
    {
      "entropy": 5.765431594848633,
      "epoch": 0.48240411229735075,
      "grad_norm": 1.25,
      "learning_rate": 0.0004983363804816433,
      "loss": 5.7026,
      "mean_token_accuracy": 0.16171906739473343,
      "num_tokens": 11156905.0,
      "step": 4880
    },
    {
      "entropy": 5.812859296798706,
      "epoch": 0.4828983788058521,
      "grad_norm": 1.25,
      "learning_rate": 0.0004983320942430256,
      "loss": 5.6916,
      "mean_token_accuracy": 0.16429043263196946,
      "num_tokens": 11169383.0,
      "step": 4885
    },
    {
      "entropy": 5.784932374954224,
      "epoch": 0.4833926453143535,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004983278025104036,
      "loss": 5.6305,
      "mean_token_accuracy": 0.17158691734075546,
      "num_tokens": 11179795.0,
      "step": 4890
    },
    {
      "entropy": 5.760914993286133,
      "epoch": 0.48388691182285487,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000498323505283883,
      "loss": 5.7569,
      "mean_token_accuracy": 0.15697767585515976,
      "num_tokens": 11190783.0,
      "step": 4895
    },
    {
      "entropy": 5.82587571144104,
      "epoch": 0.48438117833135624,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004983192025635698,
      "loss": 5.6969,
      "mean_token_accuracy": 0.15884145647287368,
      "num_tokens": 11203000.0,
      "step": 4900
    },
    {
      "entropy": 5.802449417114258,
      "epoch": 0.48487544483985767,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004983148943495693,
      "loss": 5.7132,
      "mean_token_accuracy": 0.1578467845916748,
      "num_tokens": 11213721.0,
      "step": 4905
    },
    {
      "entropy": 5.8142822742462155,
      "epoch": 0.48536971134835905,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004983105806419878,
      "loss": 5.7944,
      "mean_token_accuracy": 0.1531928800046444,
      "num_tokens": 11224490.0,
      "step": 4910
    },
    {
      "entropy": 5.801441812515259,
      "epoch": 0.4858639778568604,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004983062614409314,
      "loss": 5.6924,
      "mean_token_accuracy": 0.16117836385965348,
      "num_tokens": 11235022.0,
      "step": 4915
    },
    {
      "entropy": 5.850840520858765,
      "epoch": 0.4863582443653618,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004983019367465064,
      "loss": 5.7245,
      "mean_token_accuracy": 0.15540830790996552,
      "num_tokens": 11247485.0,
      "step": 4920
    },
    {
      "entropy": 5.75506625175476,
      "epoch": 0.48685251087386316,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000498297606558819,
      "loss": 5.6655,
      "mean_token_accuracy": 0.1630787342786789,
      "num_tokens": 11258488.0,
      "step": 4925
    },
    {
      "entropy": 5.736137628555298,
      "epoch": 0.4873467773823646,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004982932708779759,
      "loss": 5.7167,
      "mean_token_accuracy": 0.16207361966371536,
      "num_tokens": 11270786.0,
      "step": 4930
    },
    {
      "entropy": 5.90667872428894,
      "epoch": 0.48784104389086597,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004982889297040836,
      "loss": 5.7079,
      "mean_token_accuracy": 0.15878405645489693,
      "num_tokens": 11282434.0,
      "step": 4935
    },
    {
      "entropy": 5.791211462020874,
      "epoch": 0.48833531039936734,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004982845830372491,
      "loss": 5.694,
      "mean_token_accuracy": 0.1584634080529213,
      "num_tokens": 11292406.0,
      "step": 4940
    },
    {
      "entropy": 5.844303798675537,
      "epoch": 0.4888295769078687,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004982802308775791,
      "loss": 5.83,
      "mean_token_accuracy": 0.1521375820040703,
      "num_tokens": 11304491.0,
      "step": 4945
    },
    {
      "entropy": 5.850516319274902,
      "epoch": 0.4893238434163701,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000498275873225181,
      "loss": 5.7576,
      "mean_token_accuracy": 0.159835883975029,
      "num_tokens": 11317038.0,
      "step": 4950
    },
    {
      "entropy": 5.819036674499512,
      "epoch": 0.4898181099248715,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004982715100801616,
      "loss": 5.7688,
      "mean_token_accuracy": 0.1559791311621666,
      "num_tokens": 11329354.0,
      "step": 4955
    },
    {
      "entropy": 5.789994812011718,
      "epoch": 0.4903123764333729,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004982671414426285,
      "loss": 5.735,
      "mean_token_accuracy": 0.15645507723093033,
      "num_tokens": 11340508.0,
      "step": 4960
    },
    {
      "entropy": 5.792756795883179,
      "epoch": 0.49080664294187426,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004982627673126889,
      "loss": 5.7224,
      "mean_token_accuracy": 0.16280135810375213,
      "num_tokens": 11352608.0,
      "step": 4965
    },
    {
      "entropy": 5.749039602279663,
      "epoch": 0.49130090945037563,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004982583876904508,
      "loss": 5.6953,
      "mean_token_accuracy": 0.16312929540872573,
      "num_tokens": 11364089.0,
      "step": 4970
    },
    {
      "entropy": 5.7701911449432375,
      "epoch": 0.491795175958877,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004982540025760217,
      "loss": 5.6459,
      "mean_token_accuracy": 0.16562199592590332,
      "num_tokens": 11375521.0,
      "step": 4975
    },
    {
      "entropy": 5.782875490188599,
      "epoch": 0.49228944246737844,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004982496119695094,
      "loss": 5.7708,
      "mean_token_accuracy": 0.1639150306582451,
      "num_tokens": 11387721.0,
      "step": 4980
    },
    {
      "entropy": 5.809487771987915,
      "epoch": 0.4927837089758798,
      "grad_norm": 1.453125,
      "learning_rate": 0.000498245215871022,
      "loss": 5.7793,
      "mean_token_accuracy": 0.15731133073568343,
      "num_tokens": 11398688.0,
      "step": 4985
    },
    {
      "entropy": 5.778379106521607,
      "epoch": 0.4932779754843812,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004982408142806678,
      "loss": 5.6689,
      "mean_token_accuracy": 0.1573247194290161,
      "num_tokens": 11410420.0,
      "step": 4990
    },
    {
      "entropy": 5.766946649551391,
      "epoch": 0.49377224199288255,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004982364071985549,
      "loss": 5.6466,
      "mean_token_accuracy": 0.16381097137928008,
      "num_tokens": 11420521.0,
      "step": 4995
    },
    {
      "entropy": 5.80004506111145,
      "epoch": 0.4942665085013839,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004982319946247917,
      "loss": 5.7457,
      "mean_token_accuracy": 0.15540773868560792,
      "num_tokens": 11432550.0,
      "step": 5000
    },
    {
      "entropy": 5.876774835586548,
      "epoch": 0.49476077500988536,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004982275765594869,
      "loss": 5.7119,
      "mean_token_accuracy": 0.15528473407030105,
      "num_tokens": 11444602.0,
      "step": 5005
    },
    {
      "entropy": 5.723411178588867,
      "epoch": 0.49525504151838673,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004982231530027489,
      "loss": 5.5624,
      "mean_token_accuracy": 0.16830452531576157,
      "num_tokens": 11456187.0,
      "step": 5010
    },
    {
      "entropy": 5.714942979812622,
      "epoch": 0.4957493080268881,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004982187239546868,
      "loss": 5.7802,
      "mean_token_accuracy": 0.15819563418626786,
      "num_tokens": 11468615.0,
      "step": 5015
    },
    {
      "entropy": 5.8007875919342045,
      "epoch": 0.4962435745353895,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004982142894154095,
      "loss": 5.6822,
      "mean_token_accuracy": 0.1621628761291504,
      "num_tokens": 11480865.0,
      "step": 5020
    },
    {
      "entropy": 5.8547523021698,
      "epoch": 0.49673784104389085,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000498209849385026,
      "loss": 5.7576,
      "mean_token_accuracy": 0.1487893909215927,
      "num_tokens": 11492590.0,
      "step": 5025
    },
    {
      "entropy": 5.832790374755859,
      "epoch": 0.4972321075523923,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004982054038636456,
      "loss": 5.7379,
      "mean_token_accuracy": 0.16455563455820083,
      "num_tokens": 11503481.0,
      "step": 5030
    },
    {
      "entropy": 5.797842264175415,
      "epoch": 0.49772637406089365,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004982009528513776,
      "loss": 5.677,
      "mean_token_accuracy": 0.15781579166650772,
      "num_tokens": 11515083.0,
      "step": 5035
    },
    {
      "entropy": 5.786930131912231,
      "epoch": 0.498220640569395,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004981964963483316,
      "loss": 5.7479,
      "mean_token_accuracy": 0.15487340539693834,
      "num_tokens": 11527923.0,
      "step": 5040
    },
    {
      "entropy": 5.790785455703736,
      "epoch": 0.4987149070778964,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004981920343546171,
      "loss": 5.767,
      "mean_token_accuracy": 0.15837057381868364,
      "num_tokens": 11539456.0,
      "step": 5045
    },
    {
      "entropy": 5.834515476226807,
      "epoch": 0.49920917358639777,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000498187566870344,
      "loss": 5.6746,
      "mean_token_accuracy": 0.16797788292169571,
      "num_tokens": 11550809.0,
      "step": 5050
    },
    {
      "entropy": 5.907103157043457,
      "epoch": 0.49970344009489914,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004981830938956221,
      "loss": 5.8069,
      "mean_token_accuracy": 0.15619237422943116,
      "num_tokens": 11562545.0,
      "step": 5055
    },
    {
      "entropy": 5.729175996780396,
      "epoch": 0.5001977066034006,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004981786154305615,
      "loss": 5.6898,
      "mean_token_accuracy": 0.16165309697389602,
      "num_tokens": 11574479.0,
      "step": 5060
    },
    {
      "entropy": 5.8273388862609865,
      "epoch": 0.500691973111902,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004981741314752722,
      "loss": 5.6308,
      "mean_token_accuracy": 0.16577657908201218,
      "num_tokens": 11583942.0,
      "step": 5065
    },
    {
      "entropy": 5.768800449371338,
      "epoch": 0.5011862396204033,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004981696420298647,
      "loss": 5.7375,
      "mean_token_accuracy": 0.15999182015657426,
      "num_tokens": 11596745.0,
      "step": 5070
    },
    {
      "entropy": 5.69454779624939,
      "epoch": 0.5016805061289047,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004981651470944495,
      "loss": 5.6275,
      "mean_token_accuracy": 0.16875656843185424,
      "num_tokens": 11608950.0,
      "step": 5075
    },
    {
      "entropy": 5.708047723770141,
      "epoch": 0.5021747726374061,
      "grad_norm": 1.265625,
      "learning_rate": 0.000498160646669137,
      "loss": 5.7504,
      "mean_token_accuracy": 0.16267986744642257,
      "num_tokens": 11619996.0,
      "step": 5080
    },
    {
      "entropy": 5.8013287544250485,
      "epoch": 0.5026690391459074,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004981561407540379,
      "loss": 5.6509,
      "mean_token_accuracy": 0.16587038934230805,
      "num_tokens": 11632167.0,
      "step": 5085
    },
    {
      "entropy": 5.7611579418182375,
      "epoch": 0.5031633056544088,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004981516293492632,
      "loss": 5.6612,
      "mean_token_accuracy": 0.16275765299797057,
      "num_tokens": 11643722.0,
      "step": 5090
    },
    {
      "entropy": 5.723428201675415,
      "epoch": 0.5036575721629103,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004981471124549237,
      "loss": 5.6624,
      "mean_token_accuracy": 0.1652078226208687,
      "num_tokens": 11654890.0,
      "step": 5095
    },
    {
      "entropy": 5.806342506408692,
      "epoch": 0.5041518386714117,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004981425900711308,
      "loss": 5.7349,
      "mean_token_accuracy": 0.15667447149753572,
      "num_tokens": 11665485.0,
      "step": 5100
    },
    {
      "entropy": 5.781207275390625,
      "epoch": 0.504646105179913,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004981380621979954,
      "loss": 5.6345,
      "mean_token_accuracy": 0.1660120278596878,
      "num_tokens": 11676145.0,
      "step": 5105
    },
    {
      "entropy": 5.761055755615234,
      "epoch": 0.5051403716884144,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004981335288356291,
      "loss": 5.6519,
      "mean_token_accuracy": 0.15547278001904488,
      "num_tokens": 11687777.0,
      "step": 5110
    },
    {
      "entropy": 5.841087675094604,
      "epoch": 0.5056346381969158,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004981289899841434,
      "loss": 5.6841,
      "mean_token_accuracy": 0.1652728870511055,
      "num_tokens": 11699703.0,
      "step": 5115
    },
    {
      "entropy": 5.83178071975708,
      "epoch": 0.5061289047054172,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004981244456436499,
      "loss": 5.7897,
      "mean_token_accuracy": 0.14936670213937758,
      "num_tokens": 11711632.0,
      "step": 5120
    },
    {
      "entropy": 5.860979509353638,
      "epoch": 0.5066231712139185,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004981198958142604,
      "loss": 5.7706,
      "mean_token_accuracy": 0.15854335874319075,
      "num_tokens": 11723763.0,
      "step": 5125
    },
    {
      "entropy": 5.8356544971466064,
      "epoch": 0.5071174377224199,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004981153404960869,
      "loss": 5.7765,
      "mean_token_accuracy": 0.15582619458436966,
      "num_tokens": 11736061.0,
      "step": 5130
    },
    {
      "entropy": 5.736947250366211,
      "epoch": 0.5076117042309213,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004981107796892414,
      "loss": 5.7768,
      "mean_token_accuracy": 0.16211810559034348,
      "num_tokens": 11749342.0,
      "step": 5135
    },
    {
      "entropy": 5.870418214797974,
      "epoch": 0.5081059707394227,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004981062133938361,
      "loss": 5.7008,
      "mean_token_accuracy": 0.16118012517690658,
      "num_tokens": 11760441.0,
      "step": 5140
    },
    {
      "entropy": 5.736885452270508,
      "epoch": 0.508600237247924,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004981016416099834,
      "loss": 5.617,
      "mean_token_accuracy": 0.16642038524150848,
      "num_tokens": 11772191.0,
      "step": 5145
    },
    {
      "entropy": 5.8572736263275145,
      "epoch": 0.5090945037564255,
      "grad_norm": 1.25,
      "learning_rate": 0.0004980970643377956,
      "loss": 5.8781,
      "mean_token_accuracy": 0.14796603471040726,
      "num_tokens": 11786347.0,
      "step": 5150
    },
    {
      "entropy": 5.814970922470093,
      "epoch": 0.5095887702649269,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004980924815773855,
      "loss": 5.6264,
      "mean_token_accuracy": 0.16511821299791335,
      "num_tokens": 11797741.0,
      "step": 5155
    },
    {
      "entropy": 5.809056663513184,
      "epoch": 0.5100830367734283,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004980878933288657,
      "loss": 5.7587,
      "mean_token_accuracy": 0.15323774218559266,
      "num_tokens": 11809728.0,
      "step": 5160
    },
    {
      "entropy": 5.859042739868164,
      "epoch": 0.5105773032819296,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000498083299592349,
      "loss": 5.7098,
      "mean_token_accuracy": 0.1520225413143635,
      "num_tokens": 11822000.0,
      "step": 5165
    },
    {
      "entropy": 5.8006284713745115,
      "epoch": 0.511071569790431,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004980787003679486,
      "loss": 5.7056,
      "mean_token_accuracy": 0.15795185714960097,
      "num_tokens": 11834059.0,
      "step": 5170
    },
    {
      "entropy": 5.692893600463867,
      "epoch": 0.5115658362989324,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004980740956557776,
      "loss": 5.5905,
      "mean_token_accuracy": 0.1717328056693077,
      "num_tokens": 11844001.0,
      "step": 5175
    },
    {
      "entropy": 5.73483099937439,
      "epoch": 0.5120601028074337,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004980694854559493,
      "loss": 5.7625,
      "mean_token_accuracy": 0.16032130271196365,
      "num_tokens": 11855991.0,
      "step": 5180
    },
    {
      "entropy": 5.868142175674438,
      "epoch": 0.5125543693159351,
      "grad_norm": 1.25,
      "learning_rate": 0.000498064869768577,
      "loss": 5.7355,
      "mean_token_accuracy": 0.1608465313911438,
      "num_tokens": 11867975.0,
      "step": 5185
    },
    {
      "entropy": 5.761822128295899,
      "epoch": 0.5130486358244365,
      "grad_norm": 1.25,
      "learning_rate": 0.0004980602485937743,
      "loss": 5.6842,
      "mean_token_accuracy": 0.1614299662411213,
      "num_tokens": 11880034.0,
      "step": 5190
    },
    {
      "entropy": 5.815242528915405,
      "epoch": 0.5135429023329379,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004980556219316549,
      "loss": 5.7557,
      "mean_token_accuracy": 0.1584977239370346,
      "num_tokens": 11891490.0,
      "step": 5195
    },
    {
      "entropy": 5.83032603263855,
      "epoch": 0.5140371688414394,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004980509897823326,
      "loss": 5.6579,
      "mean_token_accuracy": 0.16577705144882202,
      "num_tokens": 11902988.0,
      "step": 5200
    },
    {
      "entropy": 5.741867351531982,
      "epoch": 0.5145314353499407,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004980463521459212,
      "loss": 5.7216,
      "mean_token_accuracy": 0.16082934737205506,
      "num_tokens": 11915088.0,
      "step": 5205
    },
    {
      "entropy": 5.786735105514526,
      "epoch": 0.5150257018584421,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004980417090225351,
      "loss": 5.7334,
      "mean_token_accuracy": 0.16337004601955413,
      "num_tokens": 11926789.0,
      "step": 5210
    },
    {
      "entropy": 5.9148640632629395,
      "epoch": 0.5155199683669435,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004980370604122883,
      "loss": 5.7611,
      "mean_token_accuracy": 0.15934159755706787,
      "num_tokens": 11938876.0,
      "step": 5215
    },
    {
      "entropy": 5.8335541725158695,
      "epoch": 0.5160142348754448,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004980324063152952,
      "loss": 5.7172,
      "mean_token_accuracy": 0.16266890615224838,
      "num_tokens": 11950099.0,
      "step": 5220
    },
    {
      "entropy": 5.842185688018799,
      "epoch": 0.5165085013839462,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004980277467316704,
      "loss": 5.7757,
      "mean_token_accuracy": 0.15510211884975433,
      "num_tokens": 11962585.0,
      "step": 5225
    },
    {
      "entropy": 5.820865440368652,
      "epoch": 0.5170027678924476,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004980230816615282,
      "loss": 5.7006,
      "mean_token_accuracy": 0.16341910213232042,
      "num_tokens": 11972579.0,
      "step": 5230
    },
    {
      "entropy": 5.80379729270935,
      "epoch": 0.517497034400949,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004980184111049837,
      "loss": 5.6335,
      "mean_token_accuracy": 0.16957199275493623,
      "num_tokens": 11983337.0,
      "step": 5235
    },
    {
      "entropy": 5.778374338150025,
      "epoch": 0.5179913009094503,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004980137350621518,
      "loss": 5.7555,
      "mean_token_accuracy": 0.1562938116490841,
      "num_tokens": 11995086.0,
      "step": 5240
    },
    {
      "entropy": 5.847301006317139,
      "epoch": 0.5184855674179517,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004980090535331473,
      "loss": 5.7757,
      "mean_token_accuracy": 0.1603434130549431,
      "num_tokens": 12007814.0,
      "step": 5245
    },
    {
      "entropy": 5.7690938949584964,
      "epoch": 0.5189798339264532,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004980043665180857,
      "loss": 5.6201,
      "mean_token_accuracy": 0.16814614981412887,
      "num_tokens": 12020714.0,
      "step": 5250
    },
    {
      "entropy": 5.707731676101685,
      "epoch": 0.5194741004349546,
      "grad_norm": 1.5,
      "learning_rate": 0.0004979996740170819,
      "loss": 5.6433,
      "mean_token_accuracy": 0.1608419969677925,
      "num_tokens": 12031994.0,
      "step": 5255
    },
    {
      "entropy": 5.841365575790405,
      "epoch": 0.5199683669434559,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004979949760302515,
      "loss": 5.682,
      "mean_token_accuracy": 0.15966767072677612,
      "num_tokens": 12042887.0,
      "step": 5260
    },
    {
      "entropy": 5.849974203109741,
      "epoch": 0.5204626334519573,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004979902725577101,
      "loss": 5.735,
      "mean_token_accuracy": 0.15892827957868577,
      "num_tokens": 12054426.0,
      "step": 5265
    },
    {
      "entropy": 5.810884571075439,
      "epoch": 0.5209568999604587,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004979855635995735,
      "loss": 5.7355,
      "mean_token_accuracy": 0.15427639484405517,
      "num_tokens": 12066360.0,
      "step": 5270
    },
    {
      "entropy": 5.898036861419678,
      "epoch": 0.5214511664689601,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004979808491559573,
      "loss": 5.7527,
      "mean_token_accuracy": 0.15606434941291808,
      "num_tokens": 12076651.0,
      "step": 5275
    },
    {
      "entropy": 5.781798648834228,
      "epoch": 0.5219454329774614,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004979761292269778,
      "loss": 5.6263,
      "mean_token_accuracy": 0.16331419348716736,
      "num_tokens": 12087522.0,
      "step": 5280
    },
    {
      "entropy": 5.740968227386475,
      "epoch": 0.5224396994859628,
      "grad_norm": 1.375,
      "learning_rate": 0.0004979714038127508,
      "loss": 5.7151,
      "mean_token_accuracy": 0.15970026403665544,
      "num_tokens": 12097999.0,
      "step": 5285
    },
    {
      "entropy": 5.736458444595337,
      "epoch": 0.5229339659944642,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004979666729133927,
      "loss": 5.6423,
      "mean_token_accuracy": 0.15874699205160142,
      "num_tokens": 12108080.0,
      "step": 5290
    },
    {
      "entropy": 5.913213014602661,
      "epoch": 0.5234282325029656,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004979619365290199,
      "loss": 5.8909,
      "mean_token_accuracy": 0.1450125604867935,
      "num_tokens": 12122705.0,
      "step": 5295
    },
    {
      "entropy": 5.923708438873291,
      "epoch": 0.523922499011467,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004979571946597488,
      "loss": 5.7064,
      "mean_token_accuracy": 0.1583932265639305,
      "num_tokens": 12134000.0,
      "step": 5300
    },
    {
      "entropy": 5.741547060012818,
      "epoch": 0.5244167655199684,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004979524473056961,
      "loss": 5.7182,
      "mean_token_accuracy": 0.1550974190235138,
      "num_tokens": 12144825.0,
      "step": 5305
    },
    {
      "entropy": 5.817003345489502,
      "epoch": 0.5249110320284698,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004979476944669787,
      "loss": 5.8085,
      "mean_token_accuracy": 0.15489696711301804,
      "num_tokens": 12156590.0,
      "step": 5310
    },
    {
      "entropy": 5.874768495559692,
      "epoch": 0.5254052985369712,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004979429361437134,
      "loss": 5.6488,
      "mean_token_accuracy": 0.16562367230653763,
      "num_tokens": 12167478.0,
      "step": 5315
    },
    {
      "entropy": 5.818252611160278,
      "epoch": 0.5258995650454725,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004979381723360173,
      "loss": 5.6482,
      "mean_token_accuracy": 0.16822979003190994,
      "num_tokens": 12178728.0,
      "step": 5320
    },
    {
      "entropy": 5.6653368949890135,
      "epoch": 0.5263938315539739,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004979334030440076,
      "loss": 5.6163,
      "mean_token_accuracy": 0.15801714062690736,
      "num_tokens": 12190061.0,
      "step": 5325
    },
    {
      "entropy": 5.708331680297851,
      "epoch": 0.5268880980624753,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004979286282678015,
      "loss": 5.7284,
      "mean_token_accuracy": 0.1623130276799202,
      "num_tokens": 12200867.0,
      "step": 5330
    },
    {
      "entropy": 5.832484483718872,
      "epoch": 0.5273823645709766,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004979238480075166,
      "loss": 5.6842,
      "mean_token_accuracy": 0.16192447394132614,
      "num_tokens": 12212681.0,
      "step": 5335
    },
    {
      "entropy": 5.8057342052459715,
      "epoch": 0.527876631079478,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004979190622632705,
      "loss": 5.727,
      "mean_token_accuracy": 0.16040914058685302,
      "num_tokens": 12224279.0,
      "step": 5340
    },
    {
      "entropy": 5.689106702804565,
      "epoch": 0.5283708975879794,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004979142710351809,
      "loss": 5.5774,
      "mean_token_accuracy": 0.17663966119289398,
      "num_tokens": 12235754.0,
      "step": 5345
    },
    {
      "entropy": 5.832288217544556,
      "epoch": 0.5288651640964809,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004979094743233656,
      "loss": 5.706,
      "mean_token_accuracy": 0.16103370785713195,
      "num_tokens": 12246014.0,
      "step": 5350
    },
    {
      "entropy": 5.761579513549805,
      "epoch": 0.5293594306049823,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004979046721279426,
      "loss": 5.6665,
      "mean_token_accuracy": 0.16219642460346223,
      "num_tokens": 12257519.0,
      "step": 5355
    },
    {
      "entropy": 5.7685311794281,
      "epoch": 0.5298536971134836,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004978998644490302,
      "loss": 5.7177,
      "mean_token_accuracy": 0.15667742639780044,
      "num_tokens": 12269806.0,
      "step": 5360
    },
    {
      "entropy": 5.7736852169036865,
      "epoch": 0.530347963621985,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004978950512867465,
      "loss": 5.7213,
      "mean_token_accuracy": 0.15609590262174605,
      "num_tokens": 12281612.0,
      "step": 5365
    },
    {
      "entropy": 5.835879468917847,
      "epoch": 0.5308422301304864,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004978902326412099,
      "loss": 5.6872,
      "mean_token_accuracy": 0.159187912940979,
      "num_tokens": 12293022.0,
      "step": 5370
    },
    {
      "entropy": 5.790759038925171,
      "epoch": 0.5313364966389877,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000497885408512539,
      "loss": 5.6714,
      "mean_token_accuracy": 0.16126388013362886,
      "num_tokens": 12304414.0,
      "step": 5375
    },
    {
      "entropy": 5.893075704574585,
      "epoch": 0.5318307631474891,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004978805789008526,
      "loss": 5.8077,
      "mean_token_accuracy": 0.15097813457250595,
      "num_tokens": 12315618.0,
      "step": 5380
    },
    {
      "entropy": 5.829635524749756,
      "epoch": 0.5323250296559905,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004978757438062692,
      "loss": 5.6497,
      "mean_token_accuracy": 0.16604820936918258,
      "num_tokens": 12328026.0,
      "step": 5385
    },
    {
      "entropy": 5.793191528320312,
      "epoch": 0.5328192961644919,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004978709032289081,
      "loss": 5.7214,
      "mean_token_accuracy": 0.1559169754385948,
      "num_tokens": 12339441.0,
      "step": 5390
    },
    {
      "entropy": 5.85589632987976,
      "epoch": 0.5333135626729932,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004978660571688881,
      "loss": 5.7425,
      "mean_token_accuracy": 0.15749634355306624,
      "num_tokens": 12351005.0,
      "step": 5395
    },
    {
      "entropy": 5.811450242996216,
      "epoch": 0.5338078291814946,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004978612056263286,
      "loss": 5.7471,
      "mean_token_accuracy": 0.16507457196712494,
      "num_tokens": 12363147.0,
      "step": 5400
    },
    {
      "entropy": 5.825579690933227,
      "epoch": 0.5343020956899961,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004978563486013488,
      "loss": 5.6776,
      "mean_token_accuracy": 0.16178106069564818,
      "num_tokens": 12373446.0,
      "step": 5405
    },
    {
      "entropy": 5.7309596061706545,
      "epoch": 0.5347963621984975,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004978514860940683,
      "loss": 5.6618,
      "mean_token_accuracy": 0.16514517068862916,
      "num_tokens": 12384591.0,
      "step": 5410
    },
    {
      "entropy": 5.761965560913086,
      "epoch": 0.5352906287069988,
      "grad_norm": 1.5,
      "learning_rate": 0.0004978466181046067,
      "loss": 5.6196,
      "mean_token_accuracy": 0.16532472819089888,
      "num_tokens": 12396117.0,
      "step": 5415
    },
    {
      "entropy": 5.76508355140686,
      "epoch": 0.5357848952155002,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004978417446330837,
      "loss": 5.6999,
      "mean_token_accuracy": 0.16554254442453384,
      "num_tokens": 12407400.0,
      "step": 5420
    },
    {
      "entropy": 5.7422675609588625,
      "epoch": 0.5362791617240016,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004978368656796191,
      "loss": 5.6184,
      "mean_token_accuracy": 0.16983277797698976,
      "num_tokens": 12417629.0,
      "step": 5425
    },
    {
      "entropy": 5.7121666431427,
      "epoch": 0.536773428232503,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004978319812443331,
      "loss": 5.6461,
      "mean_token_accuracy": 0.1678358569741249,
      "num_tokens": 12428522.0,
      "step": 5430
    },
    {
      "entropy": 5.803666687011718,
      "epoch": 0.5372676947410043,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004978270913273458,
      "loss": 5.7394,
      "mean_token_accuracy": 0.15472337305545808,
      "num_tokens": 12440390.0,
      "step": 5435
    },
    {
      "entropy": 5.783153104782104,
      "epoch": 0.5377619612495057,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004978221959287775,
      "loss": 5.6473,
      "mean_token_accuracy": 0.16296303272247314,
      "num_tokens": 12451846.0,
      "step": 5440
    },
    {
      "entropy": 5.743931531906128,
      "epoch": 0.5382562277580071,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004978172950487486,
      "loss": 5.6829,
      "mean_token_accuracy": 0.16433457732200624,
      "num_tokens": 12462051.0,
      "step": 5445
    },
    {
      "entropy": 5.716442108154297,
      "epoch": 0.5387504942665085,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004978123886873797,
      "loss": 5.6008,
      "mean_token_accuracy": 0.17102793008089065,
      "num_tokens": 12472730.0,
      "step": 5450
    },
    {
      "entropy": 5.751273155212402,
      "epoch": 0.5392447607750099,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004978074768447915,
      "loss": 5.6123,
      "mean_token_accuracy": 0.17399577647447587,
      "num_tokens": 12482719.0,
      "step": 5455
    },
    {
      "entropy": 5.761351537704468,
      "epoch": 0.5397390272835113,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004978025595211048,
      "loss": 5.6517,
      "mean_token_accuracy": 0.16074512004852295,
      "num_tokens": 12492659.0,
      "step": 5460
    },
    {
      "entropy": 5.882458209991455,
      "epoch": 0.5402332937920127,
      "grad_norm": 1.375,
      "learning_rate": 0.0004977976367164404,
      "loss": 5.7174,
      "mean_token_accuracy": 0.16112817525863649,
      "num_tokens": 12504381.0,
      "step": 5465
    },
    {
      "entropy": 5.807905769348144,
      "epoch": 0.540727560300514,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004977927084309198,
      "loss": 5.5634,
      "mean_token_accuracy": 0.17260986417531968,
      "num_tokens": 12515041.0,
      "step": 5470
    },
    {
      "entropy": 5.804184818267823,
      "epoch": 0.5412218268090154,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004977877746646639,
      "loss": 5.7184,
      "mean_token_accuracy": 0.15993063598871232,
      "num_tokens": 12526313.0,
      "step": 5475
    },
    {
      "entropy": 5.80208010673523,
      "epoch": 0.5417160933175168,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004977828354177941,
      "loss": 5.807,
      "mean_token_accuracy": 0.14883915781974794,
      "num_tokens": 12538268.0,
      "step": 5480
    },
    {
      "entropy": 5.808342599868775,
      "epoch": 0.5422103598260182,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004977778906904321,
      "loss": 5.7029,
      "mean_token_accuracy": 0.1601228341460228,
      "num_tokens": 12548365.0,
      "step": 5485
    },
    {
      "entropy": 5.879845809936524,
      "epoch": 0.5427046263345195,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004977729404826995,
      "loss": 5.8156,
      "mean_token_accuracy": 0.14491160213947296,
      "num_tokens": 12560652.0,
      "step": 5490
    },
    {
      "entropy": 5.794532537460327,
      "epoch": 0.5431988928430209,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004977679847947179,
      "loss": 5.6637,
      "mean_token_accuracy": 0.16358893662691115,
      "num_tokens": 12572079.0,
      "step": 5495
    },
    {
      "entropy": 5.745615434646607,
      "epoch": 0.5436931593515223,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004977630236266093,
      "loss": 5.5743,
      "mean_token_accuracy": 0.16720821112394332,
      "num_tokens": 12582601.0,
      "step": 5500
    },
    {
      "entropy": 5.836099672317505,
      "epoch": 0.5441874258600238,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004977580569784958,
      "loss": 5.7246,
      "mean_token_accuracy": 0.15426233559846877,
      "num_tokens": 12595176.0,
      "step": 5505
    },
    {
      "entropy": 5.81144642829895,
      "epoch": 0.5446816923685251,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004977530848504994,
      "loss": 5.7258,
      "mean_token_accuracy": 0.15800877958536147,
      "num_tokens": 12608161.0,
      "step": 5510
    },
    {
      "entropy": 5.777765703201294,
      "epoch": 0.5451759588770265,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004977481072427427,
      "loss": 5.7436,
      "mean_token_accuracy": 0.15919482260942458,
      "num_tokens": 12620109.0,
      "step": 5515
    },
    {
      "entropy": 5.854056882858276,
      "epoch": 0.5456702253855279,
      "grad_norm": 1.640625,
      "learning_rate": 0.000497743124155348,
      "loss": 5.7314,
      "mean_token_accuracy": 0.15875154584646226,
      "num_tokens": 12630339.0,
      "step": 5520
    },
    {
      "entropy": 5.824705743789673,
      "epoch": 0.5461644918940293,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004977381355884379,
      "loss": 5.7727,
      "mean_token_accuracy": 0.16132542490959167,
      "num_tokens": 12642101.0,
      "step": 5525
    },
    {
      "entropy": 5.889630699157715,
      "epoch": 0.5466587584025306,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004977331415421351,
      "loss": 5.8264,
      "mean_token_accuracy": 0.14514302015304564,
      "num_tokens": 12655436.0,
      "step": 5530
    },
    {
      "entropy": 5.740127992630005,
      "epoch": 0.547153024911032,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004977281420165624,
      "loss": 5.5126,
      "mean_token_accuracy": 0.1704614356160164,
      "num_tokens": 12667100.0,
      "step": 5535
    },
    {
      "entropy": 5.814797306060791,
      "epoch": 0.5476472914195334,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000497723137011843,
      "loss": 5.7994,
      "mean_token_accuracy": 0.15522741228342057,
      "num_tokens": 12678788.0,
      "step": 5540
    },
    {
      "entropy": 5.793706703186035,
      "epoch": 0.5481415579280348,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004977181265280997,
      "loss": 5.6823,
      "mean_token_accuracy": 0.16902316808700563,
      "num_tokens": 12690980.0,
      "step": 5545
    },
    {
      "entropy": 5.889812088012695,
      "epoch": 0.5486358244365361,
      "grad_norm": 1.453125,
      "learning_rate": 0.000497713110565456,
      "loss": 5.7941,
      "mean_token_accuracy": 0.15406299531459808,
      "num_tokens": 12702642.0,
      "step": 5550
    },
    {
      "entropy": 5.710815286636352,
      "epoch": 0.5491300909450376,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004977080891240352,
      "loss": 5.5668,
      "mean_token_accuracy": 0.17032462358474731,
      "num_tokens": 12712750.0,
      "step": 5555
    },
    {
      "entropy": 5.71341609954834,
      "epoch": 0.549624357453539,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000497703062203961,
      "loss": 5.5853,
      "mean_token_accuracy": 0.16658094078302382,
      "num_tokens": 12724464.0,
      "step": 5560
    },
    {
      "entropy": 5.749512577056885,
      "epoch": 0.5501186239620404,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004976980298053567,
      "loss": 5.6705,
      "mean_token_accuracy": 0.1623353898525238,
      "num_tokens": 12735419.0,
      "step": 5565
    },
    {
      "entropy": 5.863523006439209,
      "epoch": 0.5506128904705417,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004976929919283465,
      "loss": 5.6631,
      "mean_token_accuracy": 0.16216106861829757,
      "num_tokens": 12746279.0,
      "step": 5570
    },
    {
      "entropy": 5.748651504516602,
      "epoch": 0.5511071569790431,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004976879485730541,
      "loss": 5.7698,
      "mean_token_accuracy": 0.15983541011810304,
      "num_tokens": 12758107.0,
      "step": 5575
    },
    {
      "entropy": 5.745821332931518,
      "epoch": 0.5516014234875445,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004976828997396036,
      "loss": 5.6676,
      "mean_token_accuracy": 0.17020906507968903,
      "num_tokens": 12769354.0,
      "step": 5580
    },
    {
      "entropy": 5.821145534515381,
      "epoch": 0.5520956899960459,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004976778454281193,
      "loss": 5.6384,
      "mean_token_accuracy": 0.1623901054263115,
      "num_tokens": 12780926.0,
      "step": 5585
    },
    {
      "entropy": 5.810956811904907,
      "epoch": 0.5525899565045472,
      "grad_norm": 1.5,
      "learning_rate": 0.0004976727856387253,
      "loss": 5.6478,
      "mean_token_accuracy": 0.15976598411798476,
      "num_tokens": 12792423.0,
      "step": 5590
    },
    {
      "entropy": 5.711795043945313,
      "epoch": 0.5530842230130486,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004976677203715464,
      "loss": 5.5853,
      "mean_token_accuracy": 0.17233408838510514,
      "num_tokens": 12803688.0,
      "step": 5595
    },
    {
      "entropy": 5.709937620162964,
      "epoch": 0.55357848952155,
      "grad_norm": 1.5546875,
      "learning_rate": 0.000497662649626707,
      "loss": 5.5821,
      "mean_token_accuracy": 0.16821181625127793,
      "num_tokens": 12814617.0,
      "step": 5600
    },
    {
      "entropy": 5.731007671356201,
      "epoch": 0.5540727560300513,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004976575734043318,
      "loss": 5.6118,
      "mean_token_accuracy": 0.16292986273765564,
      "num_tokens": 12826027.0,
      "step": 5605
    },
    {
      "entropy": 5.753634214401245,
      "epoch": 0.5545670225385528,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004976524917045457,
      "loss": 5.7005,
      "mean_token_accuracy": 0.15490526854991912,
      "num_tokens": 12837454.0,
      "step": 5610
    },
    {
      "entropy": 5.759040403366089,
      "epoch": 0.5550612890470542,
      "grad_norm": 1.640625,
      "learning_rate": 0.000497647404527474,
      "loss": 5.597,
      "mean_token_accuracy": 0.16277759820222854,
      "num_tokens": 12848550.0,
      "step": 5615
    },
    {
      "entropy": 5.775821399688721,
      "epoch": 0.5555555555555556,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004976423118732413,
      "loss": 5.6663,
      "mean_token_accuracy": 0.16614630222320556,
      "num_tokens": 12858270.0,
      "step": 5620
    },
    {
      "entropy": 5.715994215011596,
      "epoch": 0.556049822064057,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004976372137419733,
      "loss": 5.6312,
      "mean_token_accuracy": 0.16284454464912415,
      "num_tokens": 12869507.0,
      "step": 5625
    },
    {
      "entropy": 5.775624847412109,
      "epoch": 0.5565440885725583,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004976321101337954,
      "loss": 5.5959,
      "mean_token_accuracy": 0.16878042072057725,
      "num_tokens": 12880097.0,
      "step": 5630
    },
    {
      "entropy": 5.714468574523925,
      "epoch": 0.5570383550810597,
      "grad_norm": 1.390625,
      "learning_rate": 0.000497627001048833,
      "loss": 5.5886,
      "mean_token_accuracy": 0.1684863582253456,
      "num_tokens": 12891302.0,
      "step": 5635
    },
    {
      "entropy": 5.662768936157226,
      "epoch": 0.5575326215895611,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004976218864872118,
      "loss": 5.6089,
      "mean_token_accuracy": 0.17541296780109406,
      "num_tokens": 12903950.0,
      "step": 5640
    },
    {
      "entropy": 5.763148927688599,
      "epoch": 0.5580268880980624,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004976167664490576,
      "loss": 5.6523,
      "mean_token_accuracy": 0.16255742013454438,
      "num_tokens": 12915713.0,
      "step": 5645
    },
    {
      "entropy": 5.76856484413147,
      "epoch": 0.5585211546065638,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004976116409344966,
      "loss": 5.6154,
      "mean_token_accuracy": 0.16408945620059967,
      "num_tokens": 12926655.0,
      "step": 5650
    },
    {
      "entropy": 5.692210292816162,
      "epoch": 0.5590154211150652,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004976065099436546,
      "loss": 5.6092,
      "mean_token_accuracy": 0.16390977948904037,
      "num_tokens": 12938144.0,
      "step": 5655
    },
    {
      "entropy": 5.833284473419189,
      "epoch": 0.5595096876235667,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004976013734766579,
      "loss": 5.7924,
      "mean_token_accuracy": 0.15775922387838365,
      "num_tokens": 12950486.0,
      "step": 5660
    },
    {
      "entropy": 5.755685710906983,
      "epoch": 0.560003954132068,
      "grad_norm": 1.71875,
      "learning_rate": 0.000497596231533633,
      "loss": 5.5279,
      "mean_token_accuracy": 0.1694363996386528,
      "num_tokens": 12960966.0,
      "step": 5665
    },
    {
      "entropy": 5.704496192932129,
      "epoch": 0.5604982206405694,
      "grad_norm": 1.65625,
      "learning_rate": 0.000497591084114706,
      "loss": 5.6336,
      "mean_token_accuracy": 0.16190723776817323,
      "num_tokens": 12971960.0,
      "step": 5670
    },
    {
      "entropy": 5.789654350280761,
      "epoch": 0.5609924871490708,
      "grad_norm": 1.6875,
      "learning_rate": 0.000497585931220004,
      "loss": 5.7266,
      "mean_token_accuracy": 0.16142417639493942,
      "num_tokens": 12984186.0,
      "step": 5675
    },
    {
      "entropy": 5.7306139945983885,
      "epoch": 0.5614867536575722,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004975807728496535,
      "loss": 5.562,
      "mean_token_accuracy": 0.17325062155723572,
      "num_tokens": 12994205.0,
      "step": 5680
    },
    {
      "entropy": 5.799353694915771,
      "epoch": 0.5619810201660735,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004975756090037815,
      "loss": 5.6845,
      "mean_token_accuracy": 0.16368188560009003,
      "num_tokens": 13006143.0,
      "step": 5685
    },
    {
      "entropy": 5.88020429611206,
      "epoch": 0.5624752866745749,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000497570439682515,
      "loss": 5.7204,
      "mean_token_accuracy": 0.16078546196222304,
      "num_tokens": 13017849.0,
      "step": 5690
    },
    {
      "entropy": 5.665234613418579,
      "epoch": 0.5629695531830763,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000497565264885981,
      "loss": 5.5975,
      "mean_token_accuracy": 0.1682819738984108,
      "num_tokens": 13030301.0,
      "step": 5695
    },
    {
      "entropy": 5.71648907661438,
      "epoch": 0.5634638196915777,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004975600846143072,
      "loss": 5.6572,
      "mean_token_accuracy": 0.16545034497976302,
      "num_tokens": 13040229.0,
      "step": 5700
    },
    {
      "entropy": 5.736776065826416,
      "epoch": 0.563958086200079,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004975548988676206,
      "loss": 5.592,
      "mean_token_accuracy": 0.16608160734176636,
      "num_tokens": 13050752.0,
      "step": 5705
    },
    {
      "entropy": 5.809312772750855,
      "epoch": 0.5644523527085805,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004975497076460491,
      "loss": 5.7965,
      "mean_token_accuracy": 0.15191729068756105,
      "num_tokens": 13062003.0,
      "step": 5710
    },
    {
      "entropy": 5.793465518951416,
      "epoch": 0.5649466192170819,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004975445109497202,
      "loss": 5.708,
      "mean_token_accuracy": 0.1548692211508751,
      "num_tokens": 13073746.0,
      "step": 5715
    },
    {
      "entropy": 5.859099817276001,
      "epoch": 0.5654408857255833,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004975393087787617,
      "loss": 5.7765,
      "mean_token_accuracy": 0.15511635392904283,
      "num_tokens": 13086068.0,
      "step": 5720
    },
    {
      "entropy": 5.81207914352417,
      "epoch": 0.5659351522340846,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004975341011333019,
      "loss": 5.6729,
      "mean_token_accuracy": 0.16146194487810134,
      "num_tokens": 13097078.0,
      "step": 5725
    },
    {
      "entropy": 5.746955919265747,
      "epoch": 0.566429418742586,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004975288880134685,
      "loss": 5.6566,
      "mean_token_accuracy": 0.16354386880993843,
      "num_tokens": 13108450.0,
      "step": 5730
    },
    {
      "entropy": 5.790970754623413,
      "epoch": 0.5669236852510874,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004975236694193901,
      "loss": 5.7057,
      "mean_token_accuracy": 0.15594887360930443,
      "num_tokens": 13120216.0,
      "step": 5735
    },
    {
      "entropy": 5.838050889968872,
      "epoch": 0.5674179517595888,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004975184453511948,
      "loss": 5.6929,
      "mean_token_accuracy": 0.15619661808013915,
      "num_tokens": 13131214.0,
      "step": 5740
    },
    {
      "entropy": 5.817377376556396,
      "epoch": 0.5679122182680901,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004975132158090112,
      "loss": 5.7391,
      "mean_token_accuracy": 0.15629347264766694,
      "num_tokens": 13143792.0,
      "step": 5745
    },
    {
      "entropy": 5.72936897277832,
      "epoch": 0.5684064847765915,
      "grad_norm": 2.78125,
      "learning_rate": 0.000497507980792968,
      "loss": 5.7115,
      "mean_token_accuracy": 0.15781604796648024,
      "num_tokens": 13156022.0,
      "step": 5750
    },
    {
      "entropy": 5.779087352752685,
      "epoch": 0.5689007512850929,
      "grad_norm": 1.5703125,
      "learning_rate": 0.000497502740303194,
      "loss": 5.657,
      "mean_token_accuracy": 0.16479096114635466,
      "num_tokens": 13167374.0,
      "step": 5755
    },
    {
      "entropy": 5.718874120712281,
      "epoch": 0.5693950177935944,
      "grad_norm": 1.6640625,
      "learning_rate": 0.000497497494339818,
      "loss": 5.5917,
      "mean_token_accuracy": 0.17131200879812242,
      "num_tokens": 13177662.0,
      "step": 5760
    },
    {
      "entropy": 5.713671731948852,
      "epoch": 0.5698892843020957,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004974922429029692,
      "loss": 5.6569,
      "mean_token_accuracy": 0.16335739940404892,
      "num_tokens": 13188375.0,
      "step": 5765
    },
    {
      "entropy": 5.797548484802246,
      "epoch": 0.5703835508105971,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004974869859927767,
      "loss": 5.5823,
      "mean_token_accuracy": 0.16842106431722642,
      "num_tokens": 13198988.0,
      "step": 5770
    },
    {
      "entropy": 5.837228012084961,
      "epoch": 0.5708778173190985,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004974817236093697,
      "loss": 5.6669,
      "mean_token_accuracy": 0.15082263350486755,
      "num_tokens": 13210699.0,
      "step": 5775
    },
    {
      "entropy": 5.689919376373291,
      "epoch": 0.5713720838275999,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004974764557528779,
      "loss": 5.5902,
      "mean_token_accuracy": 0.16678860187530517,
      "num_tokens": 13221925.0,
      "step": 5780
    },
    {
      "entropy": 5.740795373916626,
      "epoch": 0.5718663503361012,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004974711824234308,
      "loss": 5.7047,
      "mean_token_accuracy": 0.15749675184488296,
      "num_tokens": 13233278.0,
      "step": 5785
    },
    {
      "entropy": 5.813934755325318,
      "epoch": 0.5723606168446026,
      "grad_norm": 1.421875,
      "learning_rate": 0.000497465903621158,
      "loss": 5.6713,
      "mean_token_accuracy": 0.16290869414806367,
      "num_tokens": 13245423.0,
      "step": 5790
    },
    {
      "entropy": 5.772157192230225,
      "epoch": 0.572854883353104,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004974606193461895,
      "loss": 5.7195,
      "mean_token_accuracy": 0.16523960530757903,
      "num_tokens": 13256312.0,
      "step": 5795
    },
    {
      "entropy": 5.7896302223205565,
      "epoch": 0.5733491498616053,
      "grad_norm": 1.625,
      "learning_rate": 0.0004974553295986553,
      "loss": 5.671,
      "mean_token_accuracy": 0.16062716990709305,
      "num_tokens": 13268158.0,
      "step": 5800
    },
    {
      "entropy": 5.740266132354736,
      "epoch": 0.5738434163701067,
      "grad_norm": 4.375,
      "learning_rate": 0.0004974500343786854,
      "loss": 5.6451,
      "mean_token_accuracy": 0.16627271771430968,
      "num_tokens": 13279625.0,
      "step": 5805
    },
    {
      "entropy": 5.76525650024414,
      "epoch": 0.5743376828786081,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004974447336864102,
      "loss": 5.666,
      "mean_token_accuracy": 0.15998833552002906,
      "num_tokens": 13291967.0,
      "step": 5810
    },
    {
      "entropy": 5.686205053329468,
      "epoch": 0.5748319493871096,
      "grad_norm": 1.609375,
      "learning_rate": 0.00049743942752196,
      "loss": 5.553,
      "mean_token_accuracy": 0.1758944123983383,
      "num_tokens": 13303641.0,
      "step": 5815
    },
    {
      "entropy": 5.69308819770813,
      "epoch": 0.575326215895611,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004974341158854654,
      "loss": 5.6128,
      "mean_token_accuracy": 0.1627964958548546,
      "num_tokens": 13314754.0,
      "step": 5820
    },
    {
      "entropy": 5.784170627593994,
      "epoch": 0.5758204824041123,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004974287987770571,
      "loss": 5.7281,
      "mean_token_accuracy": 0.1593188002705574,
      "num_tokens": 13325574.0,
      "step": 5825
    },
    {
      "entropy": 5.7553788185119625,
      "epoch": 0.5763147489126137,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004974234761968658,
      "loss": 5.6266,
      "mean_token_accuracy": 0.17157164961099625,
      "num_tokens": 13335946.0,
      "step": 5830
    },
    {
      "entropy": 5.772668933868408,
      "epoch": 0.5768090154211151,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004974181481450224,
      "loss": 5.6394,
      "mean_token_accuracy": 0.16848790645599365,
      "num_tokens": 13347812.0,
      "step": 5835
    },
    {
      "entropy": 5.708554220199585,
      "epoch": 0.5773032819296164,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004974128146216581,
      "loss": 5.5988,
      "mean_token_accuracy": 0.16473570913076402,
      "num_tokens": 13358492.0,
      "step": 5840
    },
    {
      "entropy": 5.8079423904418945,
      "epoch": 0.5777975484381178,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004974074756269042,
      "loss": 5.7656,
      "mean_token_accuracy": 0.15715205073356628,
      "num_tokens": 13370465.0,
      "step": 5845
    },
    {
      "entropy": 5.811331558227539,
      "epoch": 0.5782918149466192,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004974021311608917,
      "loss": 5.6226,
      "mean_token_accuracy": 0.16963550597429275,
      "num_tokens": 13382328.0,
      "step": 5850
    },
    {
      "entropy": 5.728660964965821,
      "epoch": 0.5787860814551206,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004973967812237523,
      "loss": 5.7362,
      "mean_token_accuracy": 0.1589743450284004,
      "num_tokens": 13393439.0,
      "step": 5855
    },
    {
      "entropy": 5.764760494232178,
      "epoch": 0.5792803479636219,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004973914258156177,
      "loss": 5.66,
      "mean_token_accuracy": 0.16472170054912566,
      "num_tokens": 13404778.0,
      "step": 5860
    },
    {
      "entropy": 5.827328157424927,
      "epoch": 0.5797746144721234,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004973860649366195,
      "loss": 5.6623,
      "mean_token_accuracy": 0.16409119665622712,
      "num_tokens": 13417031.0,
      "step": 5865
    },
    {
      "entropy": 5.772861671447754,
      "epoch": 0.5802688809806248,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004973806985868895,
      "loss": 5.6507,
      "mean_token_accuracy": 0.16444518119096757,
      "num_tokens": 13428030.0,
      "step": 5870
    },
    {
      "entropy": 5.775449323654175,
      "epoch": 0.5807631474891262,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004973753267665599,
      "loss": 5.651,
      "mean_token_accuracy": 0.1662319228053093,
      "num_tokens": 13438934.0,
      "step": 5875
    },
    {
      "entropy": 5.7868541240692135,
      "epoch": 0.5812574139976275,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004973699494757627,
      "loss": 5.6287,
      "mean_token_accuracy": 0.1601264923810959,
      "num_tokens": 13450542.0,
      "step": 5880
    },
    {
      "entropy": 5.67889986038208,
      "epoch": 0.5817516805061289,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004973645667146302,
      "loss": 5.6767,
      "mean_token_accuracy": 0.1716054067015648,
      "num_tokens": 13462382.0,
      "step": 5885
    },
    {
      "entropy": 5.7738378047943115,
      "epoch": 0.5822459470146303,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004973591784832949,
      "loss": 5.6777,
      "mean_token_accuracy": 0.15822771340608596,
      "num_tokens": 13473852.0,
      "step": 5890
    },
    {
      "entropy": 5.811481475830078,
      "epoch": 0.5827402135231317,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004973537847818894,
      "loss": 5.7087,
      "mean_token_accuracy": 0.15851057916879654,
      "num_tokens": 13485763.0,
      "step": 5895
    },
    {
      "entropy": 5.740776443481446,
      "epoch": 0.583234480031633,
      "grad_norm": 1.25,
      "learning_rate": 0.0004973483856105462,
      "loss": 5.6722,
      "mean_token_accuracy": 0.16087493374943734,
      "num_tokens": 13498154.0,
      "step": 5900
    },
    {
      "entropy": 5.851483392715454,
      "epoch": 0.5837287465401344,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004973429809693983,
      "loss": 5.7124,
      "mean_token_accuracy": 0.15585886985063552,
      "num_tokens": 13510563.0,
      "step": 5905
    },
    {
      "entropy": 5.849711275100708,
      "epoch": 0.5842230130486358,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004973375708585785,
      "loss": 5.7279,
      "mean_token_accuracy": 0.16208239942789077,
      "num_tokens": 13521987.0,
      "step": 5910
    },
    {
      "entropy": 5.787547874450683,
      "epoch": 0.5847172795571373,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004973321552782201,
      "loss": 5.6743,
      "mean_token_accuracy": 0.16171410232782363,
      "num_tokens": 13534075.0,
      "step": 5915
    },
    {
      "entropy": 5.83193187713623,
      "epoch": 0.5852115460656386,
      "grad_norm": 1.46875,
      "learning_rate": 0.000497326734228456,
      "loss": 5.8267,
      "mean_token_accuracy": 0.15176279693841935,
      "num_tokens": 13546201.0,
      "step": 5920
    },
    {
      "entropy": 5.834552907943726,
      "epoch": 0.58570581257414,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004973213077094199,
      "loss": 5.6472,
      "mean_token_accuracy": 0.16397160589694976,
      "num_tokens": 13556985.0,
      "step": 5925
    },
    {
      "entropy": 5.82445592880249,
      "epoch": 0.5862000790826414,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004973158757212451,
      "loss": 5.6773,
      "mean_token_accuracy": 0.1634679839015007,
      "num_tokens": 13568617.0,
      "step": 5930
    },
    {
      "entropy": 5.668050289154053,
      "epoch": 0.5866943455911428,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004973104382640652,
      "loss": 5.6194,
      "mean_token_accuracy": 0.16359573006629943,
      "num_tokens": 13581326.0,
      "step": 5935
    },
    {
      "entropy": 5.706488561630249,
      "epoch": 0.5871886120996441,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004973049953380141,
      "loss": 5.6349,
      "mean_token_accuracy": 0.1637824833393097,
      "num_tokens": 13591776.0,
      "step": 5940
    },
    {
      "entropy": 5.85504903793335,
      "epoch": 0.5876828786081455,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004972995469432256,
      "loss": 5.7793,
      "mean_token_accuracy": 0.15763178318738938,
      "num_tokens": 13603475.0,
      "step": 5945
    },
    {
      "entropy": 5.793098735809326,
      "epoch": 0.5881771451166469,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004972940930798338,
      "loss": 5.6696,
      "mean_token_accuracy": 0.1648968607187271,
      "num_tokens": 13615348.0,
      "step": 5950
    },
    {
      "entropy": 5.720390892028808,
      "epoch": 0.5886714116251482,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004972886337479727,
      "loss": 5.5652,
      "mean_token_accuracy": 0.16782755702733992,
      "num_tokens": 13625982.0,
      "step": 5955
    },
    {
      "entropy": 5.809796619415283,
      "epoch": 0.5891656781336496,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004972831689477768,
      "loss": 5.6988,
      "mean_token_accuracy": 0.15440213531255723,
      "num_tokens": 13637203.0,
      "step": 5960
    },
    {
      "entropy": 5.763940525054932,
      "epoch": 0.5896599446421511,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004972776986793806,
      "loss": 5.6135,
      "mean_token_accuracy": 0.1651282861828804,
      "num_tokens": 13648880.0,
      "step": 5965
    },
    {
      "entropy": 5.771618938446045,
      "epoch": 0.5901542111506525,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004972722229429184,
      "loss": 5.697,
      "mean_token_accuracy": 0.16329590529203414,
      "num_tokens": 13660266.0,
      "step": 5970
    },
    {
      "entropy": 5.889302015304565,
      "epoch": 0.5906484776591538,
      "grad_norm": 1.53125,
      "learning_rate": 0.000497266741738525,
      "loss": 5.7748,
      "mean_token_accuracy": 0.14944686591625214,
      "num_tokens": 13672063.0,
      "step": 5975
    },
    {
      "entropy": 5.85100908279419,
      "epoch": 0.5911427441676552,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004972612550663353,
      "loss": 5.6693,
      "mean_token_accuracy": 0.16710522919893264,
      "num_tokens": 13683379.0,
      "step": 5980
    },
    {
      "entropy": 5.740947961807251,
      "epoch": 0.5916370106761566,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004972557629264842,
      "loss": 5.6176,
      "mean_token_accuracy": 0.16787842363119126,
      "num_tokens": 13694512.0,
      "step": 5985
    },
    {
      "entropy": 5.738776826858521,
      "epoch": 0.592131277184658,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004972502653191069,
      "loss": 5.6258,
      "mean_token_accuracy": 0.15922403559088708,
      "num_tokens": 13705642.0,
      "step": 5990
    },
    {
      "entropy": 5.705231809616089,
      "epoch": 0.5926255436931593,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004972447622443387,
      "loss": 5.5422,
      "mean_token_accuracy": 0.1716584861278534,
      "num_tokens": 13717367.0,
      "step": 5995
    },
    {
      "entropy": 5.697733211517334,
      "epoch": 0.5931198102016607,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004972392537023147,
      "loss": 5.6232,
      "mean_token_accuracy": 0.16564393490552903,
      "num_tokens": 13728577.0,
      "step": 6000
    },
    {
      "epoch": 0.5931198102016607,
      "eval_entropy": 5.626806823852464,
      "eval_loss": 5.705826759338379,
      "eval_mean_token_accuracy": 0.16681483873288716,
      "eval_num_tokens": 13728577.0,
      "eval_runtime": 26.5814,
      "eval_samples_per_second": 1223.15,
      "eval_steps_per_second": 152.927,
      "step": 6000
    }
  ],
  "logging_steps": 5,
  "max_steps": 101150,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 3000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 2.1479895263232e+16,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}